When and why we choose SP_EL0 as stack pointer? - arm

There is a field named "SpSel" in PSTATE of AArch64. And it is use to select SP between SP_EL0 or SP_ELn in EL1(or higher).
So I want to ask "Under what circumstances do we need to choose SP_EL0 as stack pointer".
In Programmer's Guide, it has sentences as following:
However, the software can switch to use SP_EL0 inside the handler. When you use this mechanism, it facilitates access to the values from the thread in the
handler.
But I can't understand it. Please help me about that.
Thanks.

Related

How to create FMU slave and initialise FMU in C using Modelica's fmi headers

I'm creating a simple FMI demo system to try out FMI where I have 1 simulator connected to an FMU which computes the state of the system (represented as a number calculated from a closed-form equation) and another FMU that controls the system via a parameter in the closed-form equation. So the system looks something like
FMU-system <--> Simulator <--> FMU-control
In every iteration, I'm updating the system state based on 1 equation, and passing it to the control, which returns a parameter to be passed to the system.
I'm using FMI 2.0.3, and have read the specification. Right now I have 3 files, 1 to act as a simulator and 2 to act as the FMUs. But I'm having difficulties with the implementation of the FMUs and the initialisation of the simulator.
To initialise the FMU, my understanding is I need to call fmi2Instantiate which has this signature.
fmi2Component fmi2Instantiate(fmi2String instanceName, fmi2Type fmuType, fmi2String fmuGUID, fmi2String fmuResourceLocation, const fmi2CallbackFunctions* functions, fmi2Boolean visible, fmi2Boolean loggingOn);
But I don't know what to pass in the function for the GUID, resource location and callback function. How should I implement the callback function and initialisation?
Then to implement the FMU, my understanding is I need to implement fmi2SetReal, fmi2GetReal and fmi2DoStep, but I can't figure out how to implement them in terms of code. These are the signatures
fmi2Status setReal(fmi2Component c, fmi2ValueReference vr[], size_t nvr, fmi2Real value[])
fmi2Status getReal(fmi2Component c, fmi2ValueReference vr[], size_t nvr, fmi2Real value[])
fmi2Status doStep(fmi2Component c, fmi2Real currentCommunicationPoint, fmi2Real communicationStepSize, fmi2Boolean noSetFMUStatePriorToCurrentPoint)
But I can't figure out how to implement these functions. Is fmi2Component c meaningless here? And I suppose I have to do the system state computation for the FMU-system in doStep. How should I update the state and pass the code here?
Sorry if this is too many questions but I was trying to look for a tutorial too and I couldn't find any.
https://github.com/traversaro/awesome-fmi
This is a curated list of Functional Mock-up Interface (FMI) libraries, tools and resources.
There are non commercial tools available. Check them out, you will get idea to implement these functions for your application.
A good starting point to implement FMI support are the open source Reference FMUs (which recently also got a simple FMU simulator) and fmpy:
https://github.com/CATIA-Systems/FMPy
https://github.com/modelica/Reference-FMUs/tree/main/fmusim

How can I return more than one value through function in C?

I have been asked in an interview how one can return more than one value from function. I have answered saying by using pointers we can achieve(call by reference) this in C. Then he told me he is looking for some other way of returning more than one value. I said we can return a struct object but here also he didn't seem to be impressed.
I would like to know others ways to return more than one value from a function.
I have seen this questions being asked here on SO, but could not find anything C specific.
The tricky problem is that the interviewer has some solution they are particularly happy with in mind and they are likely grading you by whether you have the same clever trick as them or not.
You could just name a few ways such as you did, and still not fall upon their secret trick. And if you knew their secret trick, you could well not be impressed with it.
So in these situations, its to turn it from interview into conversation. Once you detect you're not moving towards their ego, you can avoid heading towards the intimidating "I don't know" "I give up" and instead try out the "so do you have any clever solution? Is there an in-house recipe for this at Xyz Inc?" etc.
Any glimpse at their obviously self-impressed solution and you are back on firm ground where you can talk about it and ask them if they have thought about various factors that come to mind and basically interview them.
Everyone loves a good listener, and getting them to talk about their tricks is a good way to get them to leave the interview throughly impressed with you! ;)
There are a few ways:
Return value using the return statement (as you already know)
Return via references.
Return values via the heap.
Return values via global variables.
That depends on what you consider a value. If a value is a piece of information for you, more values could be a struct of values. More values could be also passed via pointers or arrays, even a char* containing a list of (non-zero alphanumerical) values. If you consider a value to be a bit of information a single returned uint32_t may hold 32 values. You could even mess around with signals or sockets or pipes or files.
But for you do not even know the use case and the requirements it imposes on the solution, it's indeed a rather hard task to come up with the right solution (and you actually did come up with some proper solutions ...).
Return a pointer to a structure, or pack several small datatypes into one large datatype, or use global variables.
The first is probably the cleanest way to do it, the other two might have their uses in certain situations.
If we pass the address instead of the true value of the parameters.
Then whenever we refer those parameters we do it with the address.
returning a pointer to structure is the suitable answer.(Obviously, the objective of the program can decide what's the best that can be done). The interviewer might have wanted you to say 'I don't know' which would have shown your lack of confidence in the field. I think you provided good solutions, though not what he had in his mind. You could have asked him about a typical scenario where he wanted multiple values to be returned and then discuss how struct-pointer is a reasonable alternative.

OOP Best Practices When One Object Needs to Modify Another

(this is a C-like environment) Say I have two instance objects, a car and a bodyShop. The car has a color iVar and corresponding accesors. The bodyShop has a method named "paintCar" that will take in a car object and change its color.
As far as implementation, in order to get the bodyShop to actually be able to change a car object's color, I see two ways to go about it.
Use the "&" operator to pass in a pointer to the car. Then the bodyShop can either tell the car to perform some method that it has to change color, or it can use the car's accessors directly.
Pass in the car object by value, do the same sort of thing to get the color changed, then have the method return a car object with a new color. Then assign the original car object to the new car object.
Option 1 seems more straightforward to me, but I'm wondering if it is in-line with OOP best practices. In general for "maximum OOP", is the "&" operator good or bad? Or, maybe I'm completely missing a better option that would make this super OOPer. Please advise :)
Option 1 is prefered:
The bodyShop can either tell the car
to perform some method that it has to
change color, or it can use the car's
accessors directly.
Even better still...create an IPaintable interface. Have Car implement IPaintable. Have BodyShop depend on IPaintable instead of Car. The benefits of this are:
Now BodyShop can paint anything that implements IPaintable (Cars, Boats, Planes, Scooters)
BodyShop is no longer tightly coupled to Car.
BodyShop has a more testable design.
I would assume that the responsibility of the bodyShop is to modify car objects, so #1 seems like the right way to go to me. I've never used a language where the "&" operator is necessary. Normally, my bodyShop object would call car.setColor(newColor) and that would be that. This way you don't have to worry about the rest of the original car's attributes, including persistence issues - you just leave them alone.
Since you're interested in the best OOP practice, you should ignore the performance hit you get with option 2. The only things you should be interested in is do either option unnecessarily increase coupling between the two classes, is encapsulation violated and is identity preserved.
Given this, option 2 is less desirable since you can't determine which other objects are holding references to the original car or worse, contain the car. In short you violate the identity constraint since two objects in the system may have different ideas of the state of the car. You run the risk of making the overall system inconsistent.
Of-course your particular environment may avoid this but it certainly would be best practice to avoid it.
Last point, does your bodyShop object have state; behaviour and identity? I realise that you have explained only the minimum necessary but possibly the bodyShop isn't really an object.
Functional v OO approaches
As an interesting aside, option 2 would close to the approach in a functional programming environment - since state changes are not allowed, your only approach would be to create a new car if it's colour changed. That's not quite what you're suggesting but it's close.
That may sound like complete overkill but it does have some interesting implications for proving the correctness of the code and parallelism.
Option 1 wins for me. The & operator is implicit in many OO languages (like Java, Python etc). You don't use "passing by value" in that languages often - only primitive types are passed in that way.
Option 2 comes with multiple problems: You might have a collection of cars, and some function unaware of it might send a car to bodyShop for painting, receive new car in return and don't update your collection of cars. See? And from more ideologic point of view - you don't create new object each time you want to modify it in real world - why should you do so in virtual one? This will lead to confusion, because it's just counterintuitive. :-)
I am not sure what this "C-like environment" mean. In C, you need this:
int paintCar(const bodyShop_t *bs, car_t *car);
where you modify the contents pointed by car. For big struct in C, you should always pass the pointer, rather than the value to a function. So, use solution 1 (if by "&" you mean the C operator).
I too agree with the first 1. I can't say it's best practice because i'm never really sure what best practice is in other peoples minds... I can tell you that best practice in my mind is the most simple method that works for the job. I've also seen this aproach taken in the hunspell win api and other c-ish api's that i've had to use. So yea i agree with scott.
http://hunspell.sourceforge.net/
//just in-case your interested in looking at other peoples code
It depends on whether the body shop's method can fail and leave the car in an indeterminate state. In that case, you're better off operating on a copy of the car, or a copy of all relevant attributes of the car. Then, only when the operation succeeds, you copy those values to the car. So you end up assigning the new car to the old car within the body shop method. Doing this correctly is necessary for exception safety in C++, and can get nasty.
It's also possible and sometimes desirable to use the other pattern - returning a new object on modification. This is useful for interactive systems which require Undo/Redo, backtracking search, and for anything involving modelling how a system of objects evolves over time.
In addition to other optinions, option 1 lets paintCar method return a completion code that indicates if the car has changed the color successfully or there were problems with it

Questions About Defined Numbers Of a Header File

I never tried to do a GUI without a GUI designer and now I'm learning how to develop Palm OS applications with the book Palm OS Programming: The Developers Guide. And on it I have this code that is a declaration of some GUI items:
#define HelloWorldForm 1000
#define HelloWorldButtonButton 1003
#define HelloWorldMenuBar 1000
#define GoodnightMoonAlert 1101
#define FirstBeep 1010
#define SecondBeepmore 1000
I want to know some things:
I need to do this in a type of order?
Why I need to declare this numbers?
In what they are going to help me?
They have anything connected to the type of item they are?
They resource IDs. You aren't required to define such macros, but if you don't, you'll instead have to use the raw integer values when you try to refer to the UI widgets in code. For example, the typical way you'd get a pointer to a UI control would be to call:
FormType* formP = FrmGetActiveForm();
UInt16 index = FrmGetObjectIndex(formP, objectID);
ControlType* controlP = FrmGetObjectPtr(formP, index);
You'd need to get pointers to the UI widgets in order to do things such as reading their states (such as for checkboxes), changing text labels, showing or hiding them dynamically, etc.
There's no type safety between the resource IDs and what you do with the pointer you get back from FrmGetObjectPtr; it's your responsibility to keep track of which ID corresponds to which type of control (the common practice is to use descriptive names).
Those are probably "human readable shortcuts" (in the form of C macros it seems) associated with GUI elements: the said elements are probably only referenced through integers in the host system. You didn't provide us with much details to work with here.
The advantage of such technique is usually associated with easier maintenance (amongst others).
They provide a couple of benefits.
Documenting the code. Would you rather have a call like LoadForm(1000); or LoadForm(HelloWorldForm); in your source code?
They may actually be used by the file that defines the resources. If you ever need to renumber the resources (perhaps you are merging two projects that both used 1000 for different forms), you would only need to modify the value in a single place.

How do I implement exceptions with nestable try-catch-finally statement with messages in C

I'm looking to implement exceptions with nestable try-catch-finally statement with messages in C using longjmp/setjmp.
I've managed to implement try-catch-else exceptions, they are not nestable. I'm also hoping to add messages to the exceptions. Any idea how I might be able to do it?
Dave Hanson has already done a really nice package of exception macros as part of his excellent book C Interfaces and Implementations. You could either use the code wholesale or learn from his techniques. For anyone who does a fair amount of C programming, the book is worth buying—it will change the way you change about C programming, and it will show you how to do object-oriented design in C.
For nesting: a stack-frame of current try/catch blocks.
Your try will be using setjmp to save to a jmpbuffer (I guess). If you've done a try, and hence are now in the scope of a try block and hit another try then you want to preserve the existing jmpbuffer and also create a new one - Push - and when catching you are longjmp-ing back to the point of the most recent try hence you Pop the latest jmpbuffer. So I think a stack-like model make sense for nested try/catch.
For implementation, I guess the simplest apporach is to reserve an array of jmpbuffers, hence limiting your try catch depth - but keeping it simple; Push and Pop just require you to track the index in that array.
For messages and other exception contents, a reserved area for "currentException".
Exception content. Keep it simple, define an Exception struct. A char array and an int. Keeping it simple, but not too simple, reserve an array of them so that you can support chaining.
For a throw you allow
throw ( "string", errcode )
Which simply zeros the array structure and makes one entry. And
catch ( exception )
Now can look in the array and finds the first entry, and then
throwChain ( "string", errcode)
Which adds the new exception to the array (if there is room, and if not can shuffle the array according some rule such as FIFO)
But, I've got to ask, why not just use C++?
Well, you cannot really implement exceptions in C since they are not supported by the language. The best you can do is emulate them using setjmp and longjmp and some diabolically clever macros.
A quick search turns up these links that may be useful to you:
Exceptions in C
Exception handling with longjmp()

Resources