similar to: [LLVMdev] problem with using DSA for a side-effect analysis

Thanks guys! I was looking into ModRef before, but I don't think that's exactly what I need. ModRef API call getModRefInfo requires a Value to be passed to it. However, some memory location M (which is a <DSNode, offset> pair) visible to the caller can be modified in the callee without any Value in the caller actually pointing to that location. I also need to capture those... I

Andrew, there used to be some code to compute the ModRef behavior of functions using DS graphs. Does that code still exist in svn? What Zvonimir is asking is essentially that, although perhaps he needs it as DS nodes explicitly rather than via the generic queries in the AliasAnalysis interface. --Vikram http://www.cs.uiuc.edu/~vadve http://llvm.org/ On Apr 3, 2008, at 12:45 PM,

On Thu, Apr 3, 2008 at 12:29 PM, Zvonimir Rakamaric <zrakamar at gmail.com> wrote: > Currently I am using DSGraph::computeNodeMapping to do the task, but I > am not sure if that's the way to go.... I believe that is the correct function to use. Andrew

On Thu, Apr 3, 2008 at 4:53 PM, Zvonimir Rakamaric <zrakamar at gmail.com> wrote: > Thanks guys! > > I was looking into ModRef before, but I don't think that's exactly > what I need. ModRef API call getModRefInfo requires a Value to be > passed to it. However, some memory location M (which is a <DSNode, > offset> pair) visible to the caller can be modified

Thanks for your help John... Yup, I looked at Local.cpp even before I fired off my question to the mailing list. Take a peek here at line 464: https://github.com/llvm-mirror/poolalloc/blob/master/lib/DSA/Local.cpp Based on my understanding of this line, if AtomicCmpXchgInst does not return a pointer type, nothing gets merged. And in the example I posted, a pointer value is indeed not returned

Hi, though DSA isn't maintained in the current build, I need to use it for analyzing dependencies among functions. Therfore I load the EQTDDataStructures pass and retrieve the mapping from nodes int the callee to the ones in the parent DSGraph with computeNodeMapping. In most cases this works well, but for some function calls I get following error: DSGraph.cpp:1379: static void

Hi, I am not exactly sure what causes your error. But if you have a reduced test case we can try looking at it. Thanks, Arushi Hello, i wrote a little pass. It tries to identify every call- / invoke-instruction recursively and calls computeCalleeCallerMapping for each. The sample program just creates a vector of integer and adds an element to it. Do you know what's wrong? Best,

Hi all, I have yet another DSA-related question :), and I would appreciate your help. Actually, the following example generates some interesting potential issues in the LLVM IR too. Here is the example in C: #define CAS(x,y,z) __atomic_compare_exchange_n(x,&(y),z,true,0,0) int main() { int *x = 0; int y = 0; int *z = x; CAS(&z,x,&y); // if (z == x) z = &y; assert(*z ==

Hi all, I've been using LLVM in my software analysis projects for quite a few years now, and several years back I relied on results of DSA analysis in my SMACK tool for checking C programs. At some point that part of SMACK got deprecated, but now I would like to revisit it since it was working quite well. Therefore, I would like to learn what's the status of the poolalloc project, and

Hi everybody! I recently started using llvm in a project on inferring additional information about pointers based on their types, casts, etc. The following simple example is giving me a headache :): typedef struct { int a; short b, c; int d, e, f; } foo; void bar(foo f) { short s; s = f.b; } int main(void) { foo p; bar(p); } Because llvm doesn't allow structures and arrays

Hi, So, SMACK is a software verifier based around LLVM, and you can find more info (PDF, title, abstract) about our recent publication here: http://soarlab.org/2014/05/smack-decoupling-source-language-details-from-verifier-implementations/ I would appreciate if you could add it to your list of LLVM-based publications. Thanks! Best, -- Zvonimir -- http://zvonimir.info http://soarlab.org/

Thanks Bill! So should I email you again in 5 months? -- http://www.zvonimir.info On Wed, Feb 6, 2013 at 1:10 AM, Bill Wendling <wendling at apple.com> wrote: > Hi Zvonimir, > > We normally list projects that use LLVM when we do a release. The next release will be 3.3, which will probably start up in 5 months or so. > > -bw > > On Feb 3, 2013, at 9:16 PM, Zvonimir

Hi, Coming back to this, I would greatly appreciate if someone could add this publication (based on TSan) to the list. Thanks, -- Zvonimir On Tue, Jun 21, 2016 at 12:07 PM Zvonimir Rakamaric <zvonimir at cs.utah.edu> wrote: > Hi, > > We recently published a paper that leverages clang/LLVM (through > ThreadSanitizer) to dynamically detect data races in OpenMP programs: >

On 12/6/12 4:47 PM, Zvonimir Rakamaric wrote: > Hi all, > > I've been using LLVM in my software analysis projects for quite a few > years now, and several years back I relied on results of DSA analysis > in my SMACK tool for checking C programs. > > At some point that part of SMACK got deprecated, but now I would like > to revisit it since it was working quite well.

Hi, Our tool SMACK (https://github.com/smackers/smack/) relies on installing LLMV from APT packages that used to be provided here: http://llvm.org/apt/ This link has been down for several weeks at this point (I think). Do you have a rough estimate for when you will bring this back? Our users are having trouble installing SMACK due to the above problem, and so I am wondering if we should look

On Feb 6, 2013, at 7:22 AM, "Criswell, John T" <criswell at illinois.edu> wrote: > Dear All, > > We can add a link to the SMACK project on the LLVM User's page at http://llvm.org/Users.html at any time. Bill, does your comment refer to the release notes, or something else? > Just to the release notes. At least, that's when we do blurbs. :) -bw > In any

Dear All, We can add a link to the SMACK project on the LLVM User's page at http://llvm.org/Users.html at any time. Bill, does your comment refer to the release notes, or something else? In any event, if the SMACK project has a web site, I can add the URL to the User's page. Is the Github URL what you want to use, or is there a web page with more information that we should link to

I was trying to call "DSGraph::computeCalleeCallerMapping" function on all caller-calle pair. But, in one case it gives an assertion failure. This is the error message I get. Processing internal callee function rt_error opt: /localhome/ssahoo2/llvm/src/lib/Analysis/DataStructure/DataStructure.cpp:2248: static void llvm::DSGraph::computeNodeMapping(const llvm::DSNodeHandle&, const

Hi, We recently published another paper that leverages LLVM (through our SMACK software verifier and novel tool called Whoop), and this time we focus on detecting concurrency bugs in device drivers: http://soarlab.org/2015/08/ase2015-ddr/ You can find all the required info (PDF, title, abstract, etc.) at the above webpage. I would appreciate if you could add this paper to your list of

Thanks John! You would not believe this :), but literally just 5 minutes ago I saw the TypeSafety pass and it seems to be exactly what we need. So we'll try to leverage that... Best, -- Zvonimir -- http://zvonimir.info http://soarlab.org/ On Mon, Dec 15, 2014 at 10:28 AM, John Criswell <jtcriswel at gmail.com> wrote: > On 12/14/14, 4:33 PM, Zvonimir Rakamaric wrote: >>