ReactOS 0.4.7 Released (reactos.org)

jeditobe writes: OSNews reports that the latest version of ReactOS has been released: "ReactOS 0.4.7 has been released, and it contains a ton of fixes, improvements, and new features. Judging by the screenshots, ReactOS 0.4.7 can run Opera, Firefox, and Mozilla all at once, which is good news for those among us who want to use ReactOS on a more daily basis. There's also a new application manager which, as the name implies, makes it easier to install and uninstall applications, similar to how package managers on Linux work. On a lower level, ReactOS can now deal with Ext2, Ext3, Ext4, BtrFS, ReiserFS, FFS, and NFS partitions." General notes, tests, and changelog for the release can be found at their respective links. A less technical community changelog for ReactOS 0.4.7 is also available. ISO images are ready at the ReactOS Download page.

Comments

get back to be when it can do CEPH with hyper-v

[Joe_Dragon]

and no I do not want a i-scsi gateway.

Re:get back to be when it can do CEPH with hyper-v

[Joe_Dragon]

run as in an driver to be able to network mount and use the pool + tie to hyper-v vm's as there disk.

Getting better

[antimatter_16]

I've been following the ReactOS project for a while, and I find it really amazing the amount of progress they've made. Whenever they release a new version, I burn a CD and try it out on a number of old machines I have lying around, and it's improved a lot in the past couple of years. It used to not boot on a number of my machines, and it's gotten better or more stable with each release.

Why not OS/2 ??

[martiniturbide]

I was not able to find a group of developers that wants to do the same thing for OS/2 Warp :_(

Re:Why not OS/2 ??

[ArchieBunker]

Because it was sold to some other company and called eComStation.

Re:Why not OS/2 ??

[Hal_Porter]

How about FreeXWorks, i.e. a vxWorks clone.

vxWorks is objectively the best OS that has ever existed. Or ever will exist.

Re:Why not OS/2 ??

[EETech1]

Back around 2001 - 2005 I worked on an ECM that ran vxWorks / Windriver. The amount of work it could do was awesome.

It's painfully obvious when your RTOS is not doing the right thing when it's scheduling spark and fuel to a running engine.

Your simply can't wait for software sometimes.

Re:Why not OS/2 ??

[Hal_Porter]

I remember having a vxWorks system on my desk that run an FTP server which could pull files off a NOR flash array. It easily outperformed the Windows PC on my desk for TCIP/IP and filesystem access, despite running a much slower CPU and having what by modern standards was really slow flash memory (of course of the time it wasn't too bad - only slightly slower on a read than uncached main memory)

And when you traced through the code you could see why. It used zbuf TCP, all the code was Ring 0, and the most common case for a filesystem read call was only about a dozen instructions - basically see if the data is in the same buffer as last time and if so return data from there.

And you could fit the whole thing - kernel, application, filesystem, TCP/IP and even an FTP server and shell for debugging - into 64-128K.

The shell was great too - if you typed

foo ( 1, 2, "aa" )

All it did was lookup foo in the symbol table and pass the arguments too it. Which means you leaked data for strings, but that didn't matter because it only happened when debugging/troubleshooting and you could just reboot to get it back. Rebooting took a couple of seconds.

Absolutely awesome software. It was like using a 70's or 80's version of Unix but it was hard realtime too - they documented the interrupt latency for each architecture.

Downside was it had a rather high licence fee and the system I worked on had no memory protection at all. Windriver had vxVMI which offered memory protection but the place I worked at had decided it was too expensive.

vxVMI just swapped the page table when a process was switched.

Re:Why not OS/2 ??

[Wootery]

objectively the best

I'm willing to bet it doesn't compete with Windows for high-performance real-time graphics on modern GPUs. Context is everything, there is no 'objective'.

I don't know much about the real-time OS world though. What's the best the FOSS world has to offer on that front?

Re:Why not OS/2 ??

[Hal_Porter]

objectively the best

Re:Why not OS/2 ??

[The123king]

http://www.osfree.org/

Re:Why not OS/2 ??

[Wootery]

Do you have something to contribute to the discussion, or not?

Re:Why not OS/2 ??

[iggymanz]

false, it's still very much alive and well and powering point of sales systems at places near you maybe including your post office.

In other words, here's your OS/2

http://www.ecomstation.com/

Re:NT4 called

[fizzer06]

I burned the last two releases to CD only to find they won't boot on real hardware so I could run them "live". Maybe I should wait for version 1 to try that again?

Re: First post!!!

[Anonymous Coward]

The correct question is: does it run Crysis?

VMS + 1 = WNT

[tepples]

There was also a FreeVMS project a while back but they decided to restart it from scratch and pretty soon it wasn't intended as a reimplementation of VMS, but rather a new OS vaguely based off of it.

I believe that project was called ReactOS.

(ReactOS is a clone of Windows NT, whose design is heavily inspired by that of Digital's VMS.)

Re:VMS + 1 = WNT

[Dog-Cow]

Will you fucking drop that VMS crap? Some of the ideas are vaguely reminiscent of VMS, but they are worlds apart in everything that matters.

Re:VMS + 1 = WNT

[drinkypoo]

(ReactOS is a clone of Windows NT, whose design is heavily inspired by that of Digital's VMS.)

Windows NT borrowed just as much from Apollo DomainOS as it did from VMS. Some of the basic architecture is similar to VMS, but only the very basic stuff. Windows NT is "heavily inspired" by VMS in the same way that DOS was heavily inspired by CPM. They look superficially similar, and that's it.

Re:VMS + 1 = WNT

[Hal_Porter]

The idea of IRQLs came from VMS.

http://www.osronline.com/showT...

OK, gather the responses from this forum and show to him.

IRQL is derived on the hardware interrupt level (the interrupt controller
register). Historically this is the PDP/VAX-11 feature, and thus a VMS feature,
though things are going back - in modern x64 CPUs, you have CR8 register as APIC
TPR, so, once again the IRQL register is embedded to the CPU.

But it is too convinient to also implement "preemptivity suspend" as an IRQL
raise. After all, ISRs run with preemptivity suspended.

Though perhaps the name IRQL was invented for NT

https://blogs.msdn.microsoft.c...

The people who built DEC's VMS operating system also helped design the processors that DEC used, and many of them came to Microsoft and designed Windows NT, which was the basis for modern versions of Windows, including Windows XP and Windows 7. These guys wanted a way to disable (very quickly) just some of the interrupts in the system. They considered it useful to hold off interrupts from some sources while servicing interrupts from other sources.

They also realized that, just as you must acquire locks in the same order everywhere in your code to avoid deadlocks, you must also service interrupts with the same relative priority every time. It doesn't work if the clock interrupts are sometimes more important than the IDE controller's interrupts and sometimes they aren't.

Interrupts are frequently called "Interrupt ReQuests" and the priority of a specific IRQ is its Level. These letters, all run together, are IRQL.

So if you lay out all the interrupt sources in the system and create a priority for each one, or sometimes a priority for each group, you can start to do interesting things.

Consider a spinlock. Spinlocks (at least in the traditional sense) are implemented by having a processor spin in a tight loop trying to atomically modify a variable. The cache coherency hardware guarantees that only one processor can do that at a time, so lock acquisition goes only to the processor that succeeds. Other processors keep spinning until they succeed.

The processor that "owns" the lock needs to release the lock as soon as possible, as the other (waiting) processors are burning up processor time waiting to acquire the lock. So you really don't want to interrupt that processor and schedule some other thread for execution, causing all the waiters to spin until the owning thread is rescheduled.

In this situation, some operating systems encourage the owner of the spinlock to disable all interrupts so that the code can't be interrupted. (Note, too, that interrupts really need to be disabled before trying to acquire the lock, or the thread might be interrupted between acquiring the lock and disabling interrupts.)

The designers of VMS and NT decided that they didn't want to disable all interrupts just because some code somewhere acquired a spinlock. Some things shouldn't wait. TLB flushes, are a good example. So if only some interrupts are disabled while a spinlock is held, then you can still briefly interrupt the code that owns the lock for much more important tasks. Perhaps even more importantly, you can interrupt the processors which are spinning, waiting to acquire a spinlock for these important tasks, causing them to do something useful instead of just spinning.

Note that this means that every spinlock has an associated IRQL, and you have to use that IRQL consistently, or the machine will deadlock. In NT, by default, every spinlock has the same IRQL, called DISPATCH_LEVEL. DISPATCH_LEVEL means, essentially, that the interrupts which can cause a thread to stop running are disabled. (More about that later.)

Interestingly you had

Re:VMS + 1 = WNT

[jbengt]

Cutler reckoned kernel mode code should enforce strict contracts like 'no page faults above dispatch level' and the kernel should BSOD if those were broken to force people to fix the bugs. From my experience of Windows 2000, XP and 7 machines - none of which have ever BSOD'd unless the hardware is failing or a driver was buggy - he was right.

XP used to bluescreen on me every once and a while until SP2 & 3. Then it became rarer, but still happened.
I can still bluescreen my Windows 7 laptop by using the trackpad instead of a mouse when in the login screen after booting, but I believe that is a known driver issue that HP refuses to fix.

Re: First post!!!

[r00t_of_all_evils]

"How far it has come?!" Well seeing as when I started following it it was only a text mode command prompt and today it's a graphic system that can use Windows hardware drivers and actually run a lot of programs that were made for Windows XP, yeah, I would say it's come far and would also say they are making good progress on their goal. It initially started out to clone Windows NT but if you've even skimmed any of the material there you'd know that they're chasing a moving goal. As times change and technology advances, their goals advance with them. Yes, they do use a lot of WINE code, but what's wrong with that? It works and it saves them a shit ton of work having to implement it all themselves, plus when they do make changes to WINE they send the changes back upstream to the WINE project which helps to advance it as well. I fail to see where the real failure is here.

Re:It happened.

[r00t_of_all_evils]

I don't have the technical knowledge required to go about doing such a thing, however I also am interested in legacy systems. I actually collect older systems, mostly laptops from the 90's. I have a pretty good collection of DOS, Windows 3.1, and Windows 9x systems. I was also kicking around the idea of trying to implement a 386 in an FPGA, but last time I researched it it looks like the 386 is too complex to be implemented in a usable way. About the best you could do would be to implement the chipset with an FPGA to support an actual 386 chip.I wish there were more people interested in legacy systems though, it sucks watching everything I grew up learning on get lost to time and forgotten.

Wat?

[SeaFox]

ReactOS 0.4.7 can run Opera, Firefox, and Mozilla all at once...

I'm intrigued by this executable named simply "Mozilla", and wish to subscribe to your newsletter.

Re:Wat?

[ausekilis]

Now we know how Mozilla continues to collect that $375 Mil/year from Yahoo. They've got a ReactOS box doing the real work. Actually, that answers a lot of questions about them...

Re:Wat?

[bill_mcgonigle]

You must be new here. They obviously mean "Seamonkey" (for those who remember "Netscape Communicator").

Re:Wat?

[R.Mo_Robert]

You must be new here. They obviously mean "Seamonkey" (for those who remember "Netscape Communicator").

I think that's the joke. ;-) (Party like it's 2002.)

Re: First post!!!

[scdeimos]

Windows already has plenty of For Fuck's Sake.

Re: First post!!!

[jeditobe]

it does! https://www.youtube.com/watch?...

Don't want to be arrogant, but...

[m.alessandrini]

I tried it a couple of times recently when I needed support for peripherals with windows-only drivers, so one of the few cases where ReactOS would be really useful for something, and both times some parts of the system were not working or not implemented. So I think nobody needs a windows replacement that only works for running high-level applications.

Re: First post!!!

[Wootery]

At least Wine does."

Re: First post!!!

[Wootery]

Mighty impressive that it works at all, even if they're at the seconds-per-frame level of performance. Congratulations ReactOS, that's an awesome milestone.

Wonder if it's running in Direct3D 9 or Direct3D 11. Crysis supports both.

Re: First post!!!

[Dog-Cow]

Windows has supported IFS (installable filesystems) since at least NT. That no one has written drivers for other filesystems is not a limitation of Windows.

I donated to this project

[Rainwulf]

Its worth it. I hope that enough people will be dis-satisfied with the shitty UI and data harvesting of windows 10 and go to this. I wonder if it can succesfully join a windows domain?

Re:I donated to this project

[ledow]

Not a chance.

It's 20+ years behind in some areas and nowhere near being any kind of replacement for anything.

The domain stuff... literally years away for any level of completeness or reliability despite being "worked on" since 2009 and before then. Hell, you can only just about trust Samba to run as a secondary. I wouldn't want to be in charge of a Samba-only AD tree (which still needs Windows tools to manage it!). And then you may as well just use Samba and the various OS clients that let you integrate to the AD tree for authentication anyway, and who cares what OS it's on?

ReactOS is nothing more than a toy, you couldn't do anything serious with it, and you couldn't even use it as a desktop replacement even if you were willing to make all kinds of compromises.

There's no way that people are going to look at Windows 10 and ReactOS and decide to run ReactOS instead. It would be less hassle just to move everything to Raspberry Pi, that's how incomplete things are.

ReactOS needed huge amounts of development over 10 years ago, it's not received it in the meantime. Even Wine is pretty much dead in practical terms, and yet that's had tons more developer time spent on it, and nobody would use either in earnest or in preference to just running a different OS entirely.

FreeDOS was a success, but it took 25 years to get the equivalent of DOS up and running and it's still an incredibly niche product - nobody is going to be running DOS as an OS like they used to, not in this day and age. BIOS-reset disks to make closed-source firmware update tools not require an MS-DOS licence? Sure. Desktop OS again? Never.

ReactOS is a much more complex task and hasn't come anywhere near close, and is suffering an even worse fate because of that.

If you could have had the current state of ReactOS, back when Windows 2000 was appearing, then you might have had the impetus to compete and make it viable. Coming to this point 20 years too late means that they can never do anything but play catch-up. It's a toy, an "emulation" almost, a niche OS. It's never going to be able to do anything useful, certainly not to mainstream computer users.

Hell, it doesn't even have full Win16/Win32 compatibility yet, and they are already dead and buried for the most part.

People's time would be better spent either running "real" Windows in a VM (and hence development effort into QEmu, Xen, etc.) or moving whatever ancient-Windows-thing they want to use to something else entirely.

As far as I see, ReactOS is like making a Win3.1/WinNT emulator, even if that's not technically accurate. And Wine does a much better job of that (and that's even less technically accurate). And the uses of such are vanishingly small anyway.

If you want a non-Windows desktop replacement, I would suggest investing into the Linux DE's and distros. Nowadays there's no reason to be running Windows.

I speak as someone who used a Linux desktop exclusively for 10 years (while managing Windows networks with thousands of clients for a living), who licensed Crossover Office for many of those years, who still uses only LibreOffice for every document, and who has 50:50 split of Linux:Windows VM's even in my workplace (where I manage all the IT). I quite get wanting independence of Windows, but ReactOS isn't it.

Currently I'm on Windows 7 (because it came with the machine) with VMWare and dozens of VMs to allow me to do "real work" on the same machine without having to faff about. I deploy 8.1 in work (made to look identical to 7, have no problems with it at all). I trialled 10 and decided against it for now, but there's no reason I couldn't put the same changes into 10 and make it look like 7 against and carry on regardless. I write cross-platform software in C that works on all the major OS.

But with that kind of competition, ReactOS has no place at all and won't without literal decades more of developer investment.

Re:I donated to this project

[Anonymous Coward]

Which is exactly why he donated. If people invest their time, skills and money into it then there's now reason it cannot 'develop'. Prior to Windows 10 it was a great idea to have a 'free' version of windows that you can install on as many machines as you like without paying, but now that Windows 10 is 'free' it's not as necessary. Still, having an open source version of Windows is only a good thing, especially if it's bright up to scratch.

Re:I donated to this project

[Anonymous Coward]

The domain stuff...

End users don't need or care about that. Most Windows users aren't joined to a domain.

Even Wine is pretty much dead in practical terms

It is? Funny, because the Wine that I use gets updated often and is so good that I have a hard time finding stuff that won't work on it. Hell, in many instances Wine is MORE compatible with more Windows software than Windows itself.

FreeDOS was a success, but it took 25 years to get the equivalent of DOS up and running

No it didn't. FreeDOS was at parity with MS-DOS ages ago and has long since surpassed it.

they can never do anything but play catch-up

ReactOS doesn't need to play catch up. It just has to be good enough. For most people, that's Windows XP compatibility.

I think perhaps you should stop worrying about what other people are doing. You seem awfully controlling and insecure. Maybe you should try doing something productive instead of sitting around whining like a petulant child.

Re:I donated to this project

[The123king]

If anything, it's more necessary. Windows 10 is an ugly, inconsistent, buggy pile of shite that i'll never let into my own home.

Oh, and i use the ReactOS regedit on a day-to-day basis when user rights have restricted access to the one baked into Windows. Also the paint clone is pretty good.

Re: First post!!!

[drinkypoo]

I fail to see where the real failure is here.

The real failure is that it will literally never be useful. Also, WINE has literally never put out a release that wasn't absolutely chock-full of regressions. If they're going to pull new code from Wine periodically, they're going to be pulling regressions with it. They have taken on a literally impossible task. You can't ever reach full compatibility with Windows because even Microsoft can't do that. Microsoft software is full of code that Microsoft doesn't understand. That's why their specs for the DOC format say things like "do what the code does here". They have no idea what that code does! Anyone who imagines that Windows isn't also full of code that nobody who works for Microsoft understands any more is strong on imagination, but weak on sense. But there's tons of Windows software which doesn't work correctly in compatibility mode; hell, there's tons of software which won't run correctly on XP Mode on Win7 x64, and what's more, that list is different than the software which won't run correctly on XP Mode on Win7 x32. Of course, now Microsoft has moved on to Hyper-V, which only runs on 64 bit Windows, and which runs Windows XP very slowly due to the lack of "enlightened" drivers, meaning that your older software that won't work properly on Windows later than XP can't be run efficiently on your modern Windows PC.

If even Microsoft can't manage to institute compatibility between Windows and Windows, what hope do these jokers have to make ReactOS compatible with Windows?

Re: First post!!!

[Anonymous Coward]

It doesn't need to have constant, up to the minute compatibility. If it reaches 90% compatibility with Windows XP, Vista/7, then it will run almost everything that anyone cares about.

Re: First post!!!

[drinkypoo]

If it reaches 90% compatibility with Windows XP, Vista/7, then it will run almost everything that anyone cares about.

Not only will that never happen, but it will never happen in ways that make almost everything occasionally explode in your face. Windows wasn't designed, it evolved, it grew from a tiny tumor to a gigantic malignant prominence. Trying to emulate that is a fool's errand. Everyone who cares has been showered in Windows licenses since time was time, and can run Windows in a VM for Windows compatibility.

Great work!

[HalAtWork]

Congratulations ReactOS Team!

Re: First post!!!

[Anonymous Coward]

You don't and can't know that. Just look at how amazing Wine is these days. ReactOS can't be any worse than Windows. At least they have competent developers, which is more than I can say for Microsoft.

Looking at Windows 1.0 and Windows 10 side by side, they look and operate very similarly. The only true evolution along the way was at Windows 95 when they introduced the start menu. Since Windows 95, Windows really hasn't changed all that much. Some superficial stuff and lots of bug fixes, but to the user they still work pretty much the same.

Re: First post!!!

[The123king]

It's trying to clone Windows XP now.

Ande if you think it's truly a fa8ilure, go put your coding hat on and do a better job!

Re: First post!!!

[Wootery]

Very cool. The game's Vulkan calls just map across to ordinary Vulkan calls in Linux, unlike Direct3D which Wine has to reimplement.

Re:ReactOS will never catch up

[The123king]

Part of the reason WINE is so good is the fact that ReactOS and WINE share, and contribute to each other, a lot of code. Most of the cloned Windows apps bundled with WINE (taskmgr, solitaire, mspaint for example) were taken from ReactOS. If ReactOS didn't exist, WINE would be a lot worse than it is.

CP/M + 1 = 86-DOS

[tepples]

Windows NT is "heavily inspired" by VMS in the same way that DOS was heavily inspired by CPM.

According to Wikipedia's article about early DOS, this would mean that Windows NT "had a command structure and application programming interface that imitated that of Digital [Equipment Corp.'s VMS] operating system, which made it easy to port programs from the latter." What other excuse would there be to keep (say) 32 priority levels, each with two half-levels?

Re: First post!!!

[drinkypoo]

Looking at Windows 1.0 and Windows 10 side by side, they look and operate very similarly.

Microsoft was on the Motif WG and involved in the creation of OS/2, which despite its wacky mouse bindings operates essentially the same way as everything else but classic MacOS as well. If they leave one positive legacy behind, it will be the unification of the majority-expected window layout. It's downright shocking (or at least surprising) when they (or anyone else making a desktop) deviates from it.

Re:Does systemd know no boundaries?

[eneville]

Yes, well, MS didn't plagiarise BSD sockets or anything ...

Re: First post!!!

[Wootery]

Good point.