similar to: [LLVMdev] Finding Targets of a Function Pointer

Patrick Alexander Simmons wrote: > Hi all, > > I was wondering whether there's any existing code to find all possible > targets of a function pointer in LLVM, and, if not, if there were a > smarter way to go about writing something to accomplish this than the > obvious way of recursively following the def-use chains and creating an > aggregate set. I think the only

I've been attempting to write my pass (which depends on DSA and pool allocation) against the SVN trunk of LLVM and the llvm-poolalloc project. However, I was thinking it might be better to use the latest stable releases of these codebases. I know that this is the 2.5 branch for LLVM, but are there any stable releases of pool allocation? If so, would there be any disadvantage to my

What is this used for? Does DSA put anything here? If I want to add call sites here during my pass, will it confuse pool allocation if it runs afterwards? If so, is there any way I could clear it at the end of my pass? Thanks, --Patrick -- If I'm not here, I've gone out to find myself. If I get back before I return, please keep me here.

Hi, I'm trying to run the pool allocation pass through opt, and I'm running into problems. It segfaults frequently; for example, it does this when the input is a simple Hello World program: [simmon12 at apoc testcases]$ opt -load /home/vadve/simmon12/llvm/llvm/projects/llvm-poolalloc/Debug/lib/libLLVMDataStructure.so -load

On Tue, Jul 21, 2009 at 10:58 PM, Patrick Simmons<simmon12 at illinois.edu> wrote: > What is this used for?  Does DSA put anything here?  If I want to add > call sites here during my pass, will it confuse pool allocation if it > runs afterwards?  If so, is there any way I could clear it at the end of > my pass? This is used for unprocessed callsites. Clearing it between DSA

Depending on the value of dsa_pass_to_use, either EquivBUDataStructures or EQTDDataStructures is used. In the case that the top-down DSA is used, information is pushed down to nodes in callees. However, if bottom-up DSA is used, information has only been merged upwards and the nodes are not necessarily equivalent. Harmen Andrew Lenharth wrote: > On Tue, Mar 16, 2010 at 3:46 PM, Patrick

Patrick Alexander Simmons wrote: > I've been attempting to write my pass (which depends on DSA and pool > allocation) against the SVN trunk of LLVM and the llvm-poolalloc > project. However, I was thinking it might be better to use the latest > stable releases of these codebases. I know that this is the 2.5 branch > for LLVM, but are there any stable releases of pool

John Criswell wrote: > Dear All, > > I have finished updating the DSA and Poolalloc source code so that it > compiles with the LLVM 2.6 API. If you check out the LLVM 2.6 branch > (directions are in the llvmdev archives; look for the email by Tanya > Lattner about the LLVM 2.6 branch), you should be able to build mainline > DSA and Pool Allocation against it. > > If you

Thanks for all your help so far. My problem is that what I have are the pool descriptors, which I by traversing the uses of poolinit and accessing the first argument of each call. I need to find the DSNode (in the original function) to which this pool descriptor corresponds. The rub is that this pool descriptor of course does not exist except in the clone. If I call getFuncInfo(), I get a

Not sure if anybody has noticed: The PoolAlloc release source code extracted from SVN won't even build. I tried it last night, on WinXP/Cygwin, for both the LLVM-2.5 release, as well as the latest LLVM, neither would build -- with compile-time errors. With that being said, I have no idea which LLVM release the current PoolAlloc trunk is checked in with. Could anybody point this out? Or, at

I figure (hopefully correctly) that I can iterate over all pool descriptors in a program by iterating over all users of poolinit and looking at the first argument. However, once I have a pool descriptor, I need to get its corresponding DSNode in the function in which it is complete (or in the global graph if it is a global). How do I do this? Thanks, --Patrick

John Criswell wrote: > You can use the -debug-pass=Arguments option to opt to print out which > DSA passes it is using. > > -- John T. > The argument list was this: -dsa-local -dsa-stdlib -dsa-bu -dsa-eqtd -poolalloc -preverify -domtree -verify So, the last DSA pass done would appear to have been "-dsa-eqtd". Does this mean pool allocation is using TDeq and not BUeq

Hey, I think I might be having a build problem. I says here [https://agora.cs.illinois.edu/display/llvmgroup/Automatic+Pool+Allocation] to load LLVMDataStructure.so and poolalloc.so when using "opt" to perform pool allocation. However, I don't have poolalloc.so -- only poolalloc.o. Does anyone know why this might be? --Patrick

Hi LLVM dev team: I am now doing an experiment to comparing Steensgaard-style and Andersen-style pointer analysis on LLVM. Since steensgaard pointer analysis is in module "poolalloc", so I installed poolalloc release 2.6 on my machine(intel X86_64 RedHatEnterpriseLinux 5.1, gcc-4.2.4), two directories "include" and "lib" were created after installation but no

聪明陈 wrote: > Hi LLVM dev team: > I am now doing an experiment to comparing Steensgaard-style and > Andersen-style pointer analysis on LLVM. Since steensgaard pointer > analysis is in module "poolalloc", so I installed poolalloc release > 2.6 on my machine(intel X86_64 RedHatEnterpriseLinux 5.1, gcc-4.2.4), > two directories "include" and "lib" were

Hi, LLVM dev team: Thanks for your suggestion, I have done the experiment to compare the two pointer analysis(Andersen and Steensgaard) methods in LLVM, but the result was unexpected. In each test, I compare these two methods using same optimization; There are several tests, each with a different optimization. The benchmark is all the 11 C programs in CINT2000 of SPEC. In all the tests, I found

Patrick Alexander Simmons wrote: > Hi, > > I'm trying to run the pool allocation pass through opt, and I'm running > into problems. It segfaults frequently; for example, it does this when > the input is a simple Hello World program: > Can you email me the bitcode file that is causing the problem? > [simmon12 at apoc testcases]$ opt -load >

Hi, LLVM dev team: Thanks for your suggestion, I have done the experiment to compare the two pointer analysis(Andersen and Steensgaard) methods in LLVM, but the result was unexpected. In each test, I compare these two methods using same optimization; There are several tests, each with a different optimization. The benchmark is all the 11 C programs in CINT2000 of SPEC. In all the tests, I found

Hi I see when LLVM builds the CallGraph SCCs. a function calling through a function pointer is conservatively assumed to call internal and external functions. Therefore, it has an edges pointing to the externalnode2, ie. the externalnode representing outgoing calls from this module. does LLVM have any function pointer analysis capabilities in the mainline ? Thanks, -Trent

Thanks Everyone for the info. I am planning to work with DSA on llvm-2.9. Hope it is working as John mentioned. Manish On Tue, Sep 20, 2011 at 7:06 AM, John Criswell <criswell at illinois.edu>wrote: > On 9/19/11 9:12 PM, Manish Gupta wrote: > > Dear All, > > I am curious to know the reason for removal of andersen pointer analysis. > Is it because of some issues? We