Toyota's killer firmware

[IceFyre13th]

Please bring back mechanical controls........drive by wire stuff is bad......nothing wrong with cables or push rods

[King Tut]

Originally Posted by IceFyre13th

Please bring back mechanical controls........drive by wire stuff is bad......nothing wrong with cables or push rods

Amen my brother. Continue fighting the good fight and preach the greatness that is throttle cables.

[suaveflooder]

Originally Posted by Tromatic

"Firmware" means hardware. You may be able to flash a new version of the firmware, but if the hardware itself is busticated there's not much you can do.

If I were Toyota, I'd be having a lawyer make a phone call about this headline:
Toyota's killer firmware: Bad design and its consequences

Look up and you'll see that I thanked ft_sjo for that information. I studied for 12 hours yesterday, so my reading comprehension last night was

Oh well. Life goes one. Does suck that we can't touch it, though

[stugray]

Originally Posted by IceFyre13th

Please bring back mechanical controls........drive by wire stuff is bad......nothing wrong with cables or push rods

Except in the case of the Toyota Unintended Acceleration (UA), MY theory is that it WAS the mechanical parts that cause the failure.

Of all the studies of the software & ECU, not a single one even mentioned the Cruise control.

I have a Solara (Camry) and on the mechanical throttle body, there are TWO cables.
One is pulled by the pedal and the other is pulled by the cruise control actuator.

In one of the cases of UA, the driver stated that when the car took off, the gas pedal fell to the floor without being pressed.

This is EXACTLY what happens if the cruise control actuator takes over.
IT pulls the throttle open which causes the gas pedal cable to go slack.

I cannot believe they did not even consider the "secondary" system to be the culprit....

[Dezoris]

I can't wait for driver less cars so I can start hacking into ECUs and direct hot women into into garage. By then I will be a creepy old man so it will be even better.

[lazyluka]

Originally Posted by phrosty

This contains some very WTF wording that doesn't mean anything to a programmer:

As an embedded programmer this line makes complete sense to me:

Toyota missed some of the calls made via pointer, missed stack usage by library and assembly functions (about 350 in total), and missed RTOS use during task switching.

This isn't exactly my field, but I don't really see a reason to use assembly when running a RTOS. Just seems like a recipe for something to go wrong. (Obviously this depends on a number of if's, but's and other design decisions that we do not know about from that article)

I also would have thought that a smarter ECU design would be a SoC with an FPGA or custom hardware to process some of the time critical stuff in real time. This would be much better than using assembly.

[Tromatic]

If I get pulled over for speeding, this will be my defense: "Toyota's killer firmware". Not my fault.

[IceFyre13th]

Originally Posted by stugray

Except in the case of the Toyota Unintended Acceleration (UA), MY theory is that it WAS the mechanical parts that cause the failure.

Of all the studies of the software & ECU, not a single one even mentioned the Cruise control.

I have a Solara (Camry) and on the mechanical throttle body, there are TWO cables.
One is pulled by the pedal and the other is pulled by the cruise control actuator.

In one of the cases of UA, the driver stated that when the car took off, the gas pedal fell to the floor without being pressed.

This is EXACTLY what happens if the cruise control actuator takes over.
IT pulls the throttle open which causes the gas pedal cable to go slack.

I cannot believe they did not even consider the "secondary" system to be the culprit....

Cruise control = drive by wire

[phrosty]

Originally Posted by lazyluka

As an embedded programmer this line makes complete sense to me:

Clearly one of us misinterpreted it. It's ambiguous at best, technical lingo soup at worst.

missed some of the calls made via pointer

All I think this could mean is that function pointers are against their guidelines, but I can't imagine any guidelines including such a weird restriction. Maybe they mean they omitted the called functions from their test cases?

missed stack usage by library and assembly functions

They're talking about "missed stack usage" being bad, not about assembly usage being bad. My best guess is they mean unaccounted for (i.e. contributing to the 94% vs 41% stack usage number) but it's far from clear.

missed RTOS use during task switching.

Guessing they're talking about cooperative multitasking here; and possibly meaning they executed some code without accounting for how it affects something else which is highly dependent on timing. Again it's not completely clear.

[FirestormFRS]

Originally Posted by stugray

Except in the case of the Toyota Unintended Acceleration (UA), MY theory is that it WAS the mechanical parts that cause the failure.

Of all the studies of the software & ECU, not a single one even mentioned the Cruise control.

I have a Solara (Camry) and on the mechanical throttle body, there are TWO cables.
One is pulled by the pedal and the other is pulled by the cruise control actuator.

In one of the cases of UA, the driver stated that when the car took off, the gas pedal fell to the floor without being pressed.

This is EXACTLY what happens if the cruise control actuator takes over.
IT pulls the throttle open which causes the gas pedal cable to go slack.

I cannot believe they did not even consider the "secondary" system to be the culprit....

You bought a Solara...Really?

But yeah the models involved in the UA cases were all drive by wire.

[stugray]

Originally Posted by FirestormFRS

You bought a Solara...Really?

Yes I bought a Solara. I was a diehard Toyota fan after having 3 Supras & one Cressida. My requirement was front wheel drive, Auto, Traction Control, because at the time I had a commute of 2-3 hours a day in winter conditions. (Supras do NOT like snow)
I still have it and think the 99 was actually a beautiful car, had a supercharger offered by TRD, and was the pace car at the Denver Grand Prix.
Honestly one of the best cars I have ever owned.
The newer body style should have the designer publicly bitch-slapped.

[ft_sjo]

I think some of you experts should actually take the lid off as modern ecu and look inside.

[Trap63]

Originally Posted by phrosty

Reeks of expert testimony using a shotgun approach -- i.e. a bunch of "could-be"s versus a definite "this is what caused the issue".

Don't get me wrong, I have no doubt there are bugs -- programmers are far from perfect and even the best, most careful of practices lead to some bugs happening. Coding is hard

The correlative of this is that understanding someone else's code is hard -- I can look at someone's code and list off a ton of "could-be"s, when they've already taken care of the corner cases in some way that I'd only realize upon a deeper analysis.



The article was clearly not written by a programmer. I'd be curious to see the original technical document. This contains some very WTF wording that doesn't mean anything to a programmer:



It lists a number of things which are not bugs, but merely describe a lack of safety nets to catch certain very specific types of bugs:



And some things here which could be bugs but the rest of the article leads me to think they just mean a potential for bugs -- this kind of code has been deemed attention-worthy by the software industry due to being difficult to get right and being the most common areas to find bugs:

"The article was clearly not written by a programmer. I'd be curious to see the original technical document. This contains some very WTF wording that doesn't mean anything to a programmer:"

In accordance to US law rules is possible to gain access to all papers? Maybe useful to understand WHERE and HOW exactly sw bugs are so wide to cause failure for system

[Fast_Freddy]

UA = driver error

Always has been, always will be.

End of story.