Displaying 20 results from an estimated 1000 matches similar to: "[LLVMdev] inserting a fucntion call at the end of basic bloc"
2011 Apr 25
0
[LLVMdev] inserting a fucntion call at the end of basic bloc
Hi Nabila,
> i would like insert a fucntion call at the end of each basic bloc
you can't, because only terminators are allowed at the end of a basic block.
However you can try to insert the call before the terminator.
...
> i tried to see an example with the demo, i saw that it instead of calling the
> function, it repeats all the instruction that are existing in the function
>
2011 Apr 25
2
[LLVMdev] inserting a fucntion call at the end of basic bloc
2011/4/25 Duncan Sands <baldrick at free.fr>
> Hi Nabila,
>
> > i would like insert a fucntion call at the end of each basic bloc
>
> you can't, because only terminators are allowed at the end of a basic
> block.
> However you can try to insert the call before the terminator.
>
Yes, i mean before the termininator,
My problem is how to call a method
suppose
2011 Apr 25
0
[LLVMdev] inserting a fucntion call at the end of basic bloc
Hi Nabila,
> My problem is how to call a method
> suppose this fucntion
> void A(int x)
> {
> x=x+1;
>
> }
>
> should i define this function and declare it at the beginig of the module and
> create for it a basic bloc?
you can just declare the function (i.e. no need to give it a body), and call it.
You can then link with an object file that defines it. This is
2011 Apr 26
2
[LLVMdev] inserting a fucntion call at the end of basic bloc
I have defined the fucntion in another object file and linked it to the
in fact the fucntion is :
void consume(int , int * );
std::vector<Value*> int32_16_params;
int32_16_params.push_back(inValue);//inValue is ConstantInt* inValue
int32_16_params.push_back(gvar_int32_y);
CallInst* int32_16 = CallInst::Create(func_consume, int32_16_params.begin(),
int32_16_params.end(), "",
2011 Apr 26
2
[LLVMdev] inserting a fucntion call at the end of basic bloc
2011/4/26 Duncan Sands <baldrick at free.fr>
> Hi Nabila,
>
>
> Now when i tried this pass an error says:
>> Wrong type for attribute noalias
>> tail call void @consume(i32 noalias 3, i32* @y) nounwind
>>
>
> noalias is only for arguments of pointer type. You probably meant it to
> be on the second argument rather than the first. I suggest you
2011 Apr 26
0
[LLVMdev] inserting a fucntion call at the end of basic bloc
Hi Nabila,
> Now when i tried this pass an error says:
> Wrong type for attribute noalias
> tail call void @consume(i32 noalias 3, i32* @y) nounwind
>
>
> noalias is only for arguments of pointer type. You probably meant it to
> be on the second argument rather than the first. I suggest you correct
> your code that adds the attribute.
2011 Apr 26
0
[LLVMdev] inserting a fucntion call at the end of basic bloc
Hi Nabila,
> Now when i tried this pass an error says:
> Wrong type for attribute noalias
> tail call void @consume(i32 noalias 3, i32* @y) nounwind
noalias is only for arguments of pointer type. You probably meant it to
be on the second argument rather than the first. I suggest you correct
your code that adds the attribute.
Ciao, Duncan.
2011 Apr 26
2
[LLVMdev] inserting a fucntion call at the end of basic bloc
2011/4/26 Duncan Sands <baldrick at free.fr>
> Hi Nabila,
>
> Now when i tried this pass an error says:
>> Wrong type for attribute noalias
>> tail call void @consume(i32 noalias 3, i32* @y) nounwind
>>
>>
>> noalias is only for arguments of pointer type. You probably meant it
>> to
>> be on the second argument
2011 Apr 26
0
[LLVMdev] inserting a fucntion call at the end of basic bloc
Hi Nabila,
> where did the "noalias" attribute and "tail call" (rather than "call") come
> from? Are you setting these yourself or running some optimization pass after
> your pass?
>
>
> i have written a module pass and i have compiled it and then
you didn't answer my questions, you just repeated what you said before.
Ciao,
2011 Apr 26
1
[LLVMdev] inserting a fucntion call at the end of basic bloc
2011/4/26 Duncan Sands <baldrick at free.fr>
> Hi Nabila,
>
>
> where did the "noalias" attribute and "tail call" (rather than "call")
>> come
>> from? Are you setting these yourself or running some optimization pass
>> after
>> your pass?
>>
>>
>> i have written a module pass and i have compiled
2011 Apr 19
4
[LLVMdev] llvm-gcc
hi all please could u help me in these points
1- I have installed llvm-gcc 4.2 and llvm 2.8 with the synaptic
package manager(ubuntu 10.10)
every time i do this commande:
llvm-gcc -O3 -emit-llvm hello.c -c -o hello.bc
it shows to me i should install llvm-gcc4.5
when i have installed llvm-gcc 4.5 i get all the times a wrong bc file
as output, because when i try to execute it with lli it says to me
2011 May 20
3
[LLVMdev] convert a char * to a value
Hi all,
Please i need help, I have a method that takes 2 arguments with type char *:
void branchPredict(char *b1, char *b2){
---
--
}
i'm supposed to add this method, in an IR basic bloc:
to add it into a basic bloc i do:
//i: is the basic bloc
std::vector<Value*> void_43_params;
Constant* tbname = ConstantArray::get(M.getContext(),i->getNameStr() ,
true);
Constant* pbname =
2011 Apr 19
2
[LLVMdev] llvm-gcc
On 4/19/11 4:37 PM, Anton Korobeynikov wrote:
>> llvm-gcc -O3 -emit-llvm hello.c -c -o hello.bc
>> it shows to me i should install llvm-gcc4.5
>> when i have installed llvm-gcc 4.5 i get all the times a wrong bc file
>> as output, because when i try to execute it with lli it says to me
>> invalid signature.
> There is no llvm-gcc 4.5. Report this bug to Ubuntu,
2011 Apr 22
0
[LLVMdev] llvm-gcc
>> i would like to add the value of the size od a basic bloc to this
>>> global variable
>>> how to convert the size of a basic bloc to integer?
>>
>> Size = number of LLVM IR instructions there? If yes, just grab the
>> number, convert it to ConstantIn and use as an initializer / store to
>> variable.
>>
>
> There's a typo in the
2011 May 20
0
[LLVMdev] convert a char * to a value
On 5/20/11 5:46 PM, Nabila ABDESSAIED wrote:
> Hi all,
>
> Please i need help, I have a method that takes 2 arguments with type
> char *:
> void branchPredict(char *b1, char *b2){
> ---
> --
> }
> i'm supposed to add this method, in an IR basic bloc:
> to add it into a basic bloc i do:
The problem is that you are passing arrays to the function instead of
2010 Oct 01
1
I18n.with_options bloc executed TWICE
Hi, to avoid putting the whole locale scope on every I18n.t call I''m
trying to use a scope, as stated in the doc
(http://guides.rubyonrails.org/active_support_core_extensions.html#with_options)
I tried with
<%= I18n.with_options :locale => :en, :scope => ''portal.visitor_menu'' do
|i18n|%>
Try <%= i18n.t :fav_songs %>
<% end %>
But it produce
2003 Jul 21
4
generate a series of fucntion
Hi there,
I want to generate a large amount of functions,
say f=function(x,t) exp(-t[1]-t[2]*g_1(x)-t[3]*g_2(1+x))
where g_1(x) and g_2(x) are from a long list of moments, such as x, x^2,
log(x), log(1+x) .. and so on.
Any suggestions on how to do this efficiently?
thanks a lot.
x.w
2011 Apr 05
4
[LLVMdev] instrument a byte code with llvm
hi,
I'm newer in llvm
i would like to instrument a byte code with a pass and as a result i would
like to get an instrumented byte code.
i would like that a pass add a method (which calculate the number of
instructions) in the end of each block.
the instrumented code should contain in each block a method that calculate
the number of instructions
Please Help me
thank you
--
Nabila ABDESSAIED
2011 Sep 03
3
question with uniroot function
Dear all,
I have the following problem with the uniroot function. I want to find
roots for the fucntion "Fp2" which is defined as below.
Fz <- function(z){0.8*pnorm(z)+p1*pnorm(z-u1)+(0.2-p1)*pnorm(z-u2)}
Fp <- function(t){(1-Fz(abs(qnorm(1-(t/2)))))+(Fz(-abs(qnorm(1-(t/2)))))}
Fp2 <- function(t) {Fp(t)-0.8*t/alpha}
th <- uniroot(Fp2, lower =0, upper =1,
2009 Jan 08
2
VaR-Monte carlo Simulation, Historic simulation, Variance-Covariance Simulation
Dear R helpers
Suppose I have a portfolio of securities with exposure to Equity, Bonds and Forex (say $ 1000000 each).
Is there any fucntion in R that will help me calculate Value at Risk (VaR) using Monte carlo Simulation , Historic simulation and Variance - Covariance Simulation.
With regards
Maithili