I'm working on a toy compiler and I'm trying to generate llvm ir bytecode and the exe compiling it with clang.
I've a simple question: I searched on google some information about the getelementptr, but I can't understand why I've to pass two i32 or i64 consts as last two arguments.
For example:
In global scope:
#string = constant [4 x i8] c"abc\00"
In function scope:
;result now contains the value of #string
; instruction | type of result | pointer to #string | what are these two i32 0?
%result = getelementptr [4 x i8], [4 x i8]* #string, i32 0, i32 0
Related
I'm trying to replace an argument to an existing function call. The initial argument is a global string and i'm trying to replace it with a local string.
No matter what i try i get a Call parameter type does not match function signature! runtime error.
Specifically i want to replace the argument to the printf function - declare dso_local i32 #printf(i8*, ...) #1. The call to it in the code looks like this:
#message = dso_local global [12 x i8] c"Hello World\00", align 1
...
%call = call i32 (i8*, ...) #printf(i8* getelementptr inbounds ([12 x i8], [12 x i8]* #message, i64 0, i64 0))
I allocate and GEP the local replacement string as such:
%0 = alloca [12 x i8], align 1
%1 = getelementptr inbounds [12 x i8], [12 x i8]* %0, i64 0
But when i use setOperand() on %call to replace the argument with %1, i get the aforementioned error. I guess [12 x i8]* is different from i8* but i can't find a way to cast it that works.
I've tried casting with CreatePointerCast() which created this:
%2 = addrspacecast [12 x i8]* %1 to i8 addrspace(8)*
But got the same error (the type for %2 is i8 addrspace(8)* which is different from i8* i guess).
Also tries explicitly specifying the i8* type when creating the GEP:
Builder->CreateInBoundsGEP(llvm::Type::getInt8PtrTy(Ctx, 8), ...)
to no avail.
I think you're using the GEP instruction wrongly.
Remember the first index steps over the pointer of type [12 x i8]*,which is what you supplied, and the second index specifies helps compute the starting address of the first character of your string, which is probably what you missed.
I cannot find anything about this, sorry.
If I have expressions with the symbolic function f(x) and now I want to replace in these expression f(x) by its explicit form how to do it?
For example:
I have
f(x):= x^2+sin(x)
and in the differentiation
diff (%e**sqrt(f(x)*a), x,2);
I want to replace now f(x) by the expression above?
Thanks
Karl
(%i1) i: integrate(f(x)*f(4*x), x, 0, 1) $
(%i2) f(x):= x^2+sin(x) $
(%i3) ev(i, f);
1
/
[ 2 2
(%o3) I (sin(x) + x ) (sin(4 x) + 16 x ) dx
]
/
0
-- Function: ev (<expr>, <arg_1>, ..., <arg_n>)
Evaluates the expression <expr> in the environment specified by the
arguments <arg_1>, ..., <arg_n>. The arguments are switches
(Boolean flags), assignments, equations, and functions. 'ev'
returns the result (another expression) of the evaluation.
When using a model object, I call func_decl get_func_decl (unsigned i) to get the value assigned to a certain function (variable). The problem I am having is taking the output of that (which is a func_decl) and converting it to int.
For example, if the model is {x |-> 4, y |-> 12, z |-> 6}, I would like to get the actual int values of those 3 variables (4, 12 and 6).
Z3 provides the function Z3_get_numeral_int for this purpose:
Z3_bool Z3_API Z3_get_numeral_int(__in Z3_context c, __in Z3_ast v, __out int* i);
Note that the last parameter is a point to an integer that will be filled with the right value if the call succeeds (it will fail for non-numerals or numerals which are not representable as an int).
There are also other functions called Z3_get_numeral_* to obtain values of different types, e.g., uint64 etc.
This method works for constants (i.e., func_decls of arity 0). To get the entries of a (non-zero arity) function definition, the following function should be used:
Z3_func_entry Z3_API Z3_func_interp_get_entry(__in Z3_context c, __in Z3_func_interp f, unsigned i);
In the attached picture there's a symbol I don't understand. To understand additive functional dependency I need to know what the symbol means. Please advice?
It's the symbol where it says: "Suppose that X ⊇ Y and that..."
⊇ = ?
Thanks!
This symbol is used in set theory to say that X is a superset of Y, ie. All elements of Y are contained within X. Note that X could be equal to Y in this meaning. Without the underlining on the symbol Y would be a strict subset of X meaning they cannot be equal
It means "X is a superset of Y". The subset symbol is the same but flipped horizontally.
A ⊇ B -> Superset: A has the same elements as B, or more
A ⊃ B -> Proper Superset: A has B's elements and more
for example:
A ⊇ B -> {1, 2, 3} ⊇ {1, 2, 3}
A ⊃ B -> {1, 2, 3, 4} ⊃ {1, 2, 3}
Why does compiling this code:
triples( [], _,_,_)->
[];
triples( Self, X, Y, none )->
[ Result || Result = { X, Y, _} <- Self ].
report:
./simple_graph.erl:63: Warning: variable 'X' is unused
./simple_graph.erl:63: Warning: variable 'Y' is unused
./simple_graph.erl:64: Warning: variable 'X' is unused
./simple_graph.erl:64: Warning: variable 'X' shadowed in generate
./simple_graph.erl:64: Warning: variable 'Y' is unused
./simple_graph.erl:64: Warning: variable 'Y' shadowed in generate
And return wrong result: full Self.
This is because variables occurring on the LHS of generators, X and Y here, are always new unbound variables local to the comprehension. This means that they are not the same variables as the X and Y in the head of triples and, therefore, there is no implicit equality test. This similar to funs where all variables occurring in the head of a fun are alse new variables local to the fun.
This is different from most of the rest of erlang, which is why the compiler not only warns that the X and Y in the head are not used but also that the X and Y in the comprehension shadow the other variables. They are also unused anywhere in the comprehension.
An easy way to get what you want is:
[ Result || Result = {X1,Y1,_} <- Self, X =:= X1, Y =:= Y1 ]