Marcin Kozlowski

c2xml is needed. Sometimes c2xml produces errors, due to missing headers (weird?). I will just skip on malformed XML.

I got others resolved. But blocked now by DominanceFrontier ``` void vSSA::getAnalysisUsage(AnalysisUsage &AU) const { //AU.addRequired(); AU.addRequired(); } bool vSSA::runOnFunction(Function &F) { // For some reason, in the DominatorTree pass,...

Thanks @Machiry Well it seems to work with it commented out, but when it is there, this is the error ``` In file included from /usr/lib/llvm-9/include/llvm/Pass.h:363:0, from /home/mk/dr_checker/llvm_analysis/MainAnalysisPasses/RangeAnalysis/include/vSSA.h:11, from /home/mk/dr_checker/llvm_analysis/MainAnalysisPasses/RangeAnalysis/src/vSSA.cpp:14:...

Figured it out. Seem to have it working with Clang/LLVM 9 and also 10 Cc @Machiry

FYI As you can see Clang target was modified to `-target x86_64`, also had to remove some flags etc.

Bunch of Bitcode is there: ``` root@530f4f95a9a6:/dockershare/linux-5.13.9/llvm_bitcode_out# tree . |-- arch | `-- x86 | |-- boot | | |-- a20.llvm.bc | | |-- bioscall.llvm.bc | | |-- cmdline.llvm.bc |...

The question is also if this could work, clang/llvm 9 bitcode analyzed by clang/llvm 3.8 based Dr_checker

OK, @deian replied to other similar issue (https://github.com/PLSysSec/sys/issues/17) _"The path is feasible but unless the tool spits out that the attack is possible it means you can reach that block...

Thank you for taking the time to explain it in such detail. It helps to understand the project better.

FYI Here seems like another false positive. Got an uninit bug that `p_quant_data_msb[0]` could be uninitialized in a call to `calc_diff_freq()`, I don't see how however. It seems it is...