A little backstory. Altera makes FPGAs. They also have their own software to develop for them. This software runs on Windows and Linux. In theory. In practice, on Linux it can be iffy. Specifically, the device programmer acts up a lot. So i was thinking - i need Windows to program the device occasio...

hto wrote:On the other hand, ReactOS does not need to invent anything new — just to follow the trodden path

Big help it is...
One can make a universe in seven days if there is no backwards compatibility to bother about.

The biggest problem i found was opening an app and it asks for admin privalages. Now most people will just click ok but it has no way of telling you if the program is trying to mess with system files. This is why I thought soft lock with a password the user defines would work to keep system files s...

Reminds me of SharpOS.
That was an opinion-shattering little bastard.

But for ReactOS that does not feel like anything remotely beneficial in any way - slowdown, incompatibility, toolset chaos, etc.

First time i've come across ReactOS it was a command line thing after boot. If you run a GUI program, then it would jump into graphics mode and run it. I think it then jumped back to console mode when you close the app. So, What version was that? Was that a feature or an incompleteness? Why wasn't t...

Don't trouble trouble till trouble troubles you?

nicamarvin2005 wrote:What's your point?

Nothing evil, just noting that a useful Windows version can run on such a machine, so it's not impossible for ROS to be able to as well.

Further investigation: I dropped all the the serial magic and tried to find the problem with good old printfs (MachConsPutChar). It seems to loop in following block in BOOLEAN MmInitializeMemoryManager(VOID). while ((MemoryDescriptor = ArcGetMemoryDescriptor(MemoryDescriptor)) != NULL) { DPRINTM(DPR...

And nothing happens! I tried the debug build now, it gives a blank screen. There is nothing over the serial line. I checked the build in Qemu - all is well, i checked the serial line with old debug build - all is well. The garbage is also not reproducible, only blank screen. The changes are in debug...

EmuandCo wrote:Uuuuh, we dont use cygwin in RosBE ^^

Then what is RosBE/tools/cygwin1.dll for?
Anyway, whether cygwin used or not, it was what caused the interference.

Saibamen wrote:On Win7 x64 ROSBE work perfect for me

For me too now, it was just an environment glitch.

Good news is i built the newest freeldr with debug on, and will test it out tonight.

See here . That fixed it, i haven't realized cygwin can interfere with itself. Thanks and sorry for trouble. Or go to the rosbe.cmd file with a editor and check the 1st 20 lines for _ROSBE_ORIGINALPATH declarations. One is active, one is REMed out. Now do vice versa to em and save it. This might he...

I'm having troubles building ReacOS, RosBE 1.5.1.1 installs well, but when trying to build (make or make livecd), it gives a lot of this: cygwin warning: MS-DOS style path detected: obj-i386\lib\inflib_host Preferred POSIX equivalent is: obj-i386/lib/inflib_host CYGWIN environment variable option &q...

Tried running .13 on a 366Mhz Celeron with 32 Mb of RAM, and failed.
It gets to installing drivers, then hangs solid.

For the record, Windows 98 works pretty darn good on the same system.

I would have built a new one with debug, but for some weird reason ROSBE don't work on Win7x64 too well, giving path errors.
If someone could provide one, i'll test it out.