QuickTime .MOV + Toshiba + Vista = BSOD

Question Guy writes "Apple QuickTime is involved in a troubling problem that doesn't seem to be addressed by any of the major software and hardware manufacturers involved. On Toshiba machines, such as the Protege Tablet M400s, with Windows Vista installed, opening a locally stored QuickTime .MOV causes instant bluescreen. All other video functions seem to be working in other video playback types — even streaming .MOVs work — and there is little to no 'buzz' on the Net that might push any of the parties to investigate or to play nice together (Microsoft for Vista, Intel for the GMA945 chipset, Toshiba for their custom tablet software, Apple for QuickTime). Help, anyone?"

Comments

Sounds like User Error?

[Rosyna]

It says "playing a .MOV file". A .mov (MooV) file is a container format for codecs. iTunes doesn't use a "QuickTime Player" it uses QuickTime.

It almost sounds like a particular driver or something is crashing when trying to do hardware acceleration of a particular codec (like H.264). The author seems like they're shooting bullets of blame in a wild and uncontrollable manner.

Probably Vista

[slughead]

The fact that it crashed was probably Apple's "bad", but the fact that it resulted in a BSOD is obviously Vista.

Maybe this has to do with the added layer of complexity (presumably for DRM) between the kernel and video-utilizing programs... or is that just for DirectX programs?

Has apple updated QT yet?

[peragrin]

last I knew apple hadn't updated quicktime or itunes, for Vista. So people are running into problems with a heavily drm'd OS not properly running applications that weren't designed to run on it.

should i be surprised?

Re:Title error...

[ivan256]

What would be the Mac equivalent to a Tablet PC?

Certainly not Apple's fault

[evilviper]

From TFA:

or the problem could in fact be Apple's for something in the QuickTime code that's at fault.

No, it couldn't... If you're running as an unprivileged user, the software you run shouldn't possibly be able to crash your OS.

Drivers can, and bugs in the OS can. User-run programs can only (accidentally) trigger one of those... in which case, that's a DoS exploit in the system.

RALink Chipset + Ubuntu + Network Manager= Nothing

[LighterShadeOfBlack]

So my wireless USB dongle stopped working when I upgraded to Ubuntu 7.04. It seems Network Manager doesn't like the rt73usb driver, or just about any RALink driver judging by the Ubuntu Forums. Help, anyone?

What, you mean /. isn't the place to post bug reports? Could've fooled me...

Re:Title error...

[WhiteWolf666]

And the iPhone, on which you _cannot_ install software, and which has limited data connectivity options is equivalent to a full blow Tablet PC how?

I'm an Apple fan, but come on.

Re:VLC

[wellingj]

I think you are missing the point.... not much *IS* compatible with Vista.
wow... what a role reversal with Linux...
next thing you know my dad is going to make me cookies and my mom will make me build a fence...

Hastening

[SuperKendall]

Perhaps Microsoft will be in a little bit more of a hurry when the hackers figure out what is causing the BSOD from Vista and work it into the next custom malware?

Re:I think you answered your question already.

[Anonymous Coward]

We know that Apple is not the problem because an application should never be able to invoke a BSOD, no matter how poorly written.

Re:I think you answered your question already.

[Sillygates]

A userspace application should not be able to completely crash a system.

Re:Title error...

[Anonymous Coward]

The iPhone-tuned features of OS X are far more appropriate to a slate form factor than desktop OS X or Windows Brick--excuse me, Tablet Edition.

When are people going to stop insisting that there is *a* right way to do things?

The iPhone form factor would be better if you want an ultra-portable computer you can carry around with you whereever you go. The Tablet PC form factor is better if you want a bigger screen, for instance (in order of what I think is least to most convincing) if you are giving a presentation and you want to be able to note on the screen, you want to read a ebook, or if you want to take notes in class.

Both would be very useful in their domain; neither is the be-all, end-all solution. People don't use PDAs instead of laptops or laptops instead of PDAs, and there's a reason for that.

Re:Title error...

[cronot]

Others have already said it, but I'm gonna say it again anyway: The possibility that Quicktime itself is causing the BSOD is infinitesimally next to zero - no userspace application should bring the OS down directly. Quicktime obviously is triggering the BSOD, but there isn't absolutely any "strong possibility that the issue is being caused by Apple software". The issue is most likely bad drivers - and the fact that the device uses a custom tablet isn't of any help. While Apple probably could work around the problem, they are the last party he should ask for it - The first, and most likely to get his problem resolved is Toshiba. Even if they can't to anything about it directly (i.e., the problem is on Vista), they should at least be able to put some pressure on Microsoft to get it done.

I do agree, however, that the suggestion the GP gave of dumping the device is overrated.

Where's the memory dump?

[Sergeant Beavis]

Disclaimer: I USED to work at Microsoft and I now own a Mac.

Here is an example of an idiot trying to look smart. Have you bothered to have someone look at the memory dump? What was the stop code? Did you check the event logs?

The fact is, it could be ANY of the three things mentioned or NONE of them. It could be an anti virus filter driver. It could be a memory access violation in Kernel Mode memory. It could have absolutely nothing to do with Vista or QT or even the Toshiba's drivers. It could be that the author is just stupid.

I'm leaning in the direction that this author is simply ignorant but since he felt he should write an article and place blame with minimal evidence to support his claim, he falls solidly in the stupid category.

The only fact that the author has presented is that he had a BSOD when using QT on Vista on Toshiba hardware when playing a local file. That only gives you suspects. A lawyer should know better. I've had occasions where customers swore up and down that one product was causing a BSOD and the memory dump pointed squarely at another product. Rarely (on XP) did I ever see a memory dump that actually pointed the finger at Windows. More often than not, I've seen memory dumps caused by filter drivers used by anti virus.

Perhaps Mr Fishkin should write more about being a lawyer because he damn well doesn't know much about computers.

Re:I think you answered your question already.

[Anonymous Coward]

They claimed it was impossible for an application to cause a BSOD so there couldn't be anything wrong with Mozilla.

OK, so Mozilla had a resource leak. Clearly there was also a kernel or driver bug, because a BSOD is a kernel crash.

Re:I think you answered your question already.

[Anonymous Coward]

Isn't a corner stone of "stable" and "reliability" in an modern OS the inability of any app to cause failure by sucking up all resources?

"its not a bug, its a feature!"

[dknj]

um let me run this scenario by you. you're finishing up your long 226 legal brief in Microsoft word. you go to click file save, but the directory you are saving to has 400 files. windows does some internal file processing and starts eating up the final 10k of available physical memory. swapping begins to occur until your sound card that was playing your favorite midi throws an error because windows did not feed data in time. the sound card drive freaks out, and returns a failure to the midi app. the midi app, not prepared for this, starts an infinite loop eating all of your available memory. eventually you overflow the stack and your windowing system is hosed.

you shake the mouse and click repeatedly hoping to unfreeze the computer. you hit control alt delete, nothing happens. you reset the computer silently hoping word's autosave worked for once.

now, my question to you.. is the above story what you are implying is happening here? (answer this question honestly before you continue)

because that is what happened back in Windows 3.1. Since the creation of modern operating systems, we have learned to take advantage of advanced hardware and separate each application into its own memory space (see: Intro to Operating Systems at your local community college). Thus, a single application should NOT take down your entire system. If an application is causing a BSOD and there is no funky kernel-mode hardware access going on.. the fault is on the hardware or OS (to include drivers as well). Period.

If you wish to debate this, remember that I may have just found a way to compromise your system.

Re:"its not a bug, its a feature!"

[scsirob]

I do have mod points at this time, but parent is already at +5. This is a *very* insightful post indeed.

There is *NO* excuse, *NONE WHATSOEVER* that an operating system can have, to allow it to fail simply because an application does something wierd/wrong/illegal. An operating system is there to manage resources, and that includes protecting applications from another. A BSOD means a bug in the primary function of the O/S and is *CRITICAL*. It denies other applications their protection and it causes lost productivity. Stop using this O/S until the failures is explained and fixed.

You are barking up the wrong tree.

[haraldm]

Still, a user space app should never be able to produce a BSOD, even if it is pushing the limits. It's the OSes job to keep apps from going haywire, no matter what. They use that stuff in medical devices, airplanes and space travel, forget? It's like Windows should never execute (!) a mail attachment only according to the MIME type without looking at the file itself: The single largest problem with trojans and worms. A design error is a design error.

RTFM ;-)

[hummassa]

At the risk of being filtered out, try compiling and running the following program. For best results, do it on a single-processor system:

#include <sys/types.h>
#include <unistd.h>
void main(){
  for(;;){
    fork();
  }
}

I just did.

$ cd /tmp
$ xclip -o > um.c
$ gcc um.c
$ ulimit -u 200
$ ./a.out

waits some seconds, sees cpu graph in superkaramba reach 100%, waits some more, hits ^C

$

No kernel panic, no system hosing here, ok?

YES.

[hummassa]

If any client asks for more than its fair share of resources, DENY them to it. Then, and only then, this is a client's bug. If you ask for more and more of my resources and I keep giving them to you until I myself starve, it's MY problem, not yours.