Slashdot Mirror
Wind River Announces It Likes Linux After All
onecrazyfoo writes "Wind River is going to start supporting Linux in the embedded market. Pretty big news from the largest company in the embedded tools market. What makes it even more interesting is the fact that they have been very anti-Linux and outspoken about it in the past. You can read more about their announcement at LinuxDevices.com." I'm guessing this has come about because of recent changes in the company.
Somebody direct these guys to the "Supplicants" door. =)
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
Doing a 180 degree turn that fast has to be hard on the body. It's hard on the hairline when under the sheets too.
Two roads diverged in a wood, and I - I took the one the bus load of girls just went down.
Found a link off the main article showcasing products embedded with Linux and actually shipping.
Linux is stable even when a telnet to root is posted on slashdot as the 2nd post two stories in a row...
Click here to get Instant Root
They're just doing like LynuxWorks former Lynx with their Blue Cat Linux. I have a strong suspicion that WindRiver just wants to profit from some of the Linux hype, given that, apart from the price, quite frankly, their OS and tool suites are just way better than any embedded Linux suite I've ever seen/worked on/worked with.
...
It's just another company trying to jump on the Linux bandwagon. Nothing to see here folks
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
Hopefully this will mean better developer tools for embedded Linux in the future.
This is not enough.
Do you need to run an SMP kernel on the 4 slot toaster?
You are in a maze of twisted little posts, all alike.
Fuckin' A, Linux rocks -- Wasup Wind River! That's right. Bitch, make me a sammich.
From the comments so far, I will presume most Slashdotters have no experience with WindRiver.
Exercising great restraint to avoid writing anything they would likely sue me for (such as a factual tale of my experience dealing with them for over two years), I would just like to point out that we should not, in any way, consider this "good" news.
Aside from their "quality" tools (the fixing of which I can thank for giving me a reason to learn Tcl), expect to hear about a GPL violation within a few weeks. And if they handle that accusation like they handle their customers' bug reports, well, good ol' Darl may start sounding fairly reasonable to deal with.
"Fine, leave me for that flavor of the month! But dont come crawling back with your uptime goes down!" - FreeBSD
This is an actual question, wasn't Wind River one of the major contributors to the fantastic development work done on FreeBSD?
If so, how could this be _bad_ for linux in any way shape or form? Even if you don't like the company, linux-heads should wake up and realize, any company that's investing in linux and open source is GOOD, look how far FreeBSD has come from 386/BSD and FreeBSD 2.x
Error 407 - No creative sig found
No it's been proven that tweaking the kernel can finally handle 4 slots with a 120 second process. I would recommend disabling sound and SCSI in the kernel.
Two roads diverged in a wood, and I - I took the one the bus load of girls just went down.
Indeed, the VisionProbe II sounds very similar to some Abatron hardware (a BDI2000, IIRC) I was using to do kernel debugging via GDB a few years ago when working at MontaVista.
And speaking of -- there are plenty of companies which have been in the embedded Linux market longer than Wind River, and have much more Linux-friendly engineering staffs. MontaVista is one of these, Lineo another... wouldn't it make sense for your embedded Linux business to go to a company that's been focused around Linux from the start rather than just for the last few months?
Whenever I'm building something on VxWorks, I'm mainly cursing about having to work from a Win or Sun machine. No Tornado for Linux. I would be really, really pleased if they would just release a host-tools-suite for Linux. That would make building stuff for VxWorks a much more pleasant experience. Jumping straight to "embedded linux" support sounds a little like jumping the gun, actually. They've spent 20 years developing VxWorks, it seems to me the most natural and useful Linux support would be to release a Linux-native development kit for it.
Sun did a similar thing: it embraced Linux, even though doing so was to feed and support something which was cannibalizing its core business. It is now regretting that decision. Will Wind River? Alas, I suspect so. Wind River has nothing to gain by supporting Linux.
I don't think of course that they could truly extinguish Linux, but I'm sure they could make some proprietory tools or what not that would make it harder on the competition. I'd in fact expect that this is a pre-emptive strike on their competition, which has probably been gaining on their Tornado tools and WindRiver OS. I even won't put it past them to try to put a few competitors out of business, then start pushing existing embedded Linux users over to their proprietary OS.
We use uClinux which is aimed at micros without MMUs. The distro includes a couple of glibc replacements, which are optimised for size. It seems to be quite popular, and meet a range of needs. It has an active developer community and some of the kernel changes for the no MMU systems are being fed into the 2.5 kernel so will have even wider support in the future.
Peter
Maybe I didn't read the right articles, but I saw some belief that the GPL kernel might not mix well with the embedded world, and a perception of Linux being embraced by competitors, which caused them to try the other route: BSD.
I didn't see any kind of "Linux is no good" message coming from WRS. Maybe I missed some clueless sales-drone speeches?
The people you see making the pro-linux statements today: especially Fiddler, were making similar noises around the time I left, back in 1999. The attitude hasn't fundamentally changed, they just tried other avenues and they didn't work.
WRS has long had an idiology that the runtime is not the important piece, but that the tools are where the major development value is. This stems from the origin of the company as a tools provider for VRTX. As a result, they've supported multiple runtimes over the course of the company, at times runtimes provided by alternate vendors. Thus, the Linux thing isn't new.
What might be new is that the general open source movement is providing more and more sophisticated developer tools for free, such that their custom-packaging of GCC and gui project manager/debugger/etc aren't worth the boatload of cash they're used to charging for it.
-josh
Warning : toaster1 is on fire
Considering the prices for a vxworks seat I am not surprised. I am sure embedded linux at the low end has been killing them for a couple of years now.
They also changed there strategy and there "crown jewels". The vxworks source code suddenly became cheaper.
However there is a lot more to embedded development than just buying a package and putting it on. The things vxworks does do well is it provide a very configurable hardware layer which makes moving to new hardware relatively easy. Also some of there visulation tools such as windview are very good(Oh I wish there was something similar for windows) which allows you to sort out bottle necks.
However you do pay through the nose for this (and there new licensing model has made it very expensive) and for cheap targets it is just not economic.
Choose your allies carefully, it is highly unlikely you will be held accountable for the actions of your enemies
stefan@creator:~$ eject
eject: unable to eject, last error: bread failed
stefan@creator:~$
What you just asked for is a system where you can get a tested release.
Red Hat provides that. It's their distribution. If you yank out parts and something doesn't work...it's not a huge surprise. The equivalent to BSD 4.x is Debian stable or RH 7.x (which is what most people see to use for production servers running RH, due to strong maturity). Just because Linus has tagged something as "2.6-pre-foo" doesn't mean that you have a tested, complete system. That means that *he* is getting close to the point where he's ready to hand the kernel off to *Red Hat and other distro vendors* to begin testing and integration. It doesn't mean that the kernel is end-user ready if you want a tested environment.
After the kernel hits 2.6 and it's been poked at a bit and decided that there's at least a decent chance that you can use it without problems, Red Hat will put it into Rawhide. That should be considered post-development -- about alpha quality. As a matter of fact, Rawhide doesn't even have formal "releases" -- it's a working repository for integration folks -- so it might be considered pre-alpha using conventional definitions. That's when testing on a wide variety of systems begins. When RH feels confident that there aren't known problems after trying on a number of systems, they move to beta, of which Severn is their current release. This is already much further down the pipeline than you're talking about, and it's just *beta*. You may not have a tested upgrade path, but you can probably use the software. Don't cry if it doesn't work -- it hasn't passed rigorous QA, but you can probably use it on a testing system without egregious problems. *Finally*, RH does a release. RH's releases tend to be considered a bit bleeding-edge, relative to extremely conservative distributions like Debian stable (wasn't that long ago that Debian stable finally left the 2.2 kernel). After the thing's been out a bit and bugfixes have been released (kind of like businesses wait a bit after a new Windows release to let people hammer on the thing and find any dings that missed earlier examination), *then* you can consider the thing stable in a conservative sense. *Only* from release to release is there a tested upgrade path on software.
Your problem was that you were jumping the gun WRT testing by a long way. When you upgrade from FBSD 5.x, you've at least had the piece of software "integrated" into the rest of the software. It's equivalent to Rawhide, if not beta. You were yanking something that Linus hasn't even felt comfortable handing off to QA people yet and saying "gee, it doesn't work".
Other distros have their own approachs -- I just happen to know RH's better -- but the point remains that there's a hella lot more testing in the pipeline before that software should be considered something that you'd want to put on your system. How do you know you aren't going to run into filesystem corruption, or God knows what? The development releases are releases intended specifically for the use of kernel developers, and if anyone else can get good out of 'em...great. 2.6-pre *is* part of 2.5, ignoring the unusual naming. It's just denoting the fact that there's an intention to move to 2.6 soon. 2.6.0, the first release QA/integration folks should be looking at, is still months away.
Now, a lot of Linux folks like seriously riding the metal, and some folks use development kernels, even doing ballsy things like putting them on their personal workstations. However, they also don't say "Linux sucks" if something doesn't work or they wipe out their filesystem. A fair number of 'em don't realize that what they're doing is a bit risky -- it's not just using mozilla daily snapshots. There ends up with this misperception on Slashdot that folks should be able to comfortably run development releases of kernels. Folks, it's called "development" for a reason -- it isn't even alpha yet!
May we never see th