similar to: [LLVMdev] Loop exit condition analysis

On Mon, Dec 19, 2011 at 10:13 PM, Hanfeng Qin <hanfengtsin at gmail.com> wrote: > Hi all, > I am doing loop exit condition analysis. As the following sample code > segments demonstrated, > > .... > int *c = &a; > while (*c == 0); > .... > > I want to decide which variables are related to ending this spin-loop. E.g., > in above sample, c is the direct

Hi all, I got a question on loop exit condition identification based on LLVM. I made an assumption first that loop exit conditions are located in exiting basic blocks. Then I use loop->getExitingBlocks() to get all BBs. However, this policy seemed not work for combinational exit conditions, e.g. compounding conditions consists of conditions with logical predicates. *The problem is that

Hello all, I caught a problem related to pthread library. I built llvm and llvm-gcc then compiling a multithreaded benchmark written in POSIX thread library. The command "llvm-gcc -o bench bench.c -lpthread" ran well while "llvm-gcc -c bench; llvm-ld -o bench bench.o -lpthread" crashed. It told that "llvm-ld: error: Cannot find library 'pthread'". I could

Hi chenwj, According to the link, I found that during the compiling process, the option "-emit-llvm" must be added to generate immediate object *.o. [hanfeng at os-wstation02 fft]llvm-gcc -c fft.c -emit-llvm Then using llvm-ld will generate two files - a shell excutable and a llvm bytecode data both of which would run perfect. [hanfeng at os-wstation02 fft]llvm-ld -o FFT fft.o [hanfeng

> Hi chenwj, > If it use system pthread library, then it should find it in the stand search path. I also added the option "-L/usr/lib -L/usr/lib64" but it still failed. Don't know if the link below helps or not, http://lists.cs.uiuc.edu/pipermail/llvmdev/2009-November/027274.html Regards, chenwj -- Wei-Ren Chen (陳韋任) Computer Systems Lab, Institute of Information

> According to the link, I found that during the compiling process, the option "-emit-llvm" must be added to generate immediate object *.o. > [hanfeng at os-wstation02 fft]llvm-gcc -c fft.c -emit-llvm "-emit-llvm" emits object file which contains LLVM bitcode (IR). Default file name suffix is .o, but you might want to use "-o fft.bc" to make the suffix

Hi chenwj, For the first question, even if I did not specify "-o fft.bc", it still generated fft.o which is also a llvm bitcode file. For the second question, I intended to generate a native executable "FFT". However, it only generate a llvm bitcode file fft.bc and a native POSIX shell excuatable named fft. Howerver, the contents of the shell executable fft is a wrapper to call

LLVM3.8 version. https://bugs.llvm.org//show_bug.cgi?id=24233. The example can catch exception. But I use my own code for testing, CustomEHMemoryManager allocated memory address is very large, allocateCodeSection, allocateDataSection assigned address (more than 32 bit address space) is getting smaller and smaller. Cause registerEHFrames to fail. 3.9.1 or 4.0.0 version has been supported on win

For help: Llvm generated instruction calls a function (extern), the function will have a SEH exception (EXCEPTION_ACCESS_VIOLATION), But JIT can not capture the exception of the EXCEPTION_ACCESS_VIOLATION. I saw Bug 24233. EXCEPTION_ACCESS_VIOLATION exception cannot be captured after modification. How to catch EXCEPTION_ACCESS_VIOLATION exceptions on win64 ? haifeng.qin at wellintech.com

Hi David and Mats, Thanks for your explanation. If my understanding is correct, it means we don't need to consider the side-effect of malloc/free unless compiling with -ffreestanding. Because without -ffreestanding, user defined malloc/free should be compatible with std library. It makes sense to me. My point is, in std library, malloc is allowed to return null if this malloc failed. Why

Hi, there, I cannot get accurate value for calculation. for example: ld<-sqrt(1*0.05*0.95*0.05*0.95) 0.05*0.95-ld=-6.938894e-18 0.05*0.95-ld==0 is False. I met this problem in my program, how can I handle it. Thanks. xj.

Hi, there: When plot ann predicted results I need to indicate numbers of inputs for each column. V1 V2 V3 V4 1 86.2700 49.9380 30.7630 0.1327 2 89.5127 55.9707 33.7683 0.1186 3 91.1833 58.4670 34.5610 0.1134 matplot(t, pch = 1:4, type = "o", col = rainbow(ncol(t)),xlab = "No. of inputs ", ylab = "Mean of 6 Datasets",

Hi all, Partial unrolling and runtime unrolling are enabled by default in aarch64 gcc which is help to get performance better. But these two methods are enabled for only several backends in LLVM which are X86, PowerPC and R600. I don't know the history of these two kinds of unrolling, and why they are not widely used. I also want to know is, for aarch64 backend, is it intentionally to get

Hi, I was unable to find a bug report for this with a cursory search, but would like clarification if this is intended or unavoidable behaviour: ```{r} # Create example data.frames parents <- data.frame(name=c("Sarah", "Max", "Qin", "Lex"), sex=c("F", "M", "F", "M"),

dear all I got this error when I try to use libspeex in visual studio. when I include speex/speex_echo.h, it complains about unresolved external symbol _speex_echo_state_init. And when I include mdf.c into the source, it generates whole bunch of errors. Any idea? thanks so much --------------------Configuration: audio - Win32 Debug-------------------- Compiling... mdf.c C:\Documents and

Hi Mats, I think Kevin's point is malloc can return 0, if malloc/free pair is optimized way, the semantic of the original would be changed. On the other hand, malloc/free are special functions, but programmers can still define their own versions by not linking std library, so we must assume malloc/free always have side-effect like other common functions, unless we know we will link std

Hi Hal, James told me that in PassManagerBuilder.cpp, BarrierNoopPass is already used for this kind of purpose(though there's also a fixme saying it's hacking). I think it's a good idea to use this pass here. Thanks, Kevin 2015-03-11 17:05 GMT+08:00 Hal Finkel <hfinkel at anl.gov>: > ----- Original Message ----- > > From: "Kevin Qin" <kevinqindev at

Hello there: I got 13 variables in my training/target set, the first 12 variables are mixture of numerical and categorical variables. The last one is the one I need to predict, and it is a numerical variable. >train<-read.table("train.txt") >test<-read.table("test.txt") >cl<-factor(train[,13]) >pred<-knn(train, test, clk=3, prob=TRUE) >pred I got

> I think we can do such optimization with operator new, because new never returns null. This is incorrect in the case of `new (std::nothrow) ...` - the whole point of `(std::nothrow)` is to tell new that it should return NULL in case of failure, rather than throw an exception (bad_alloc). But the point here is not the actual return value, but the fact that the compiler misses that the

On 15 Dec 2016, at 07:26, haifeng.qin at wellintech.com via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > For help: > > for loop lead to stack overflow, want to actively reclaim stack memory of alloc instruction. > > How to actively reclaim stack memory ? This sounds as if you’re putting the alloca inside the loop, not in the entry basic block and reusing it. If you