similar to: AMDGPUPromoteAlloca assume 3-dims enabled?

Dear Developers, I compiled a OpenCL kernel before (on Nov. last year) like __kernel void g(__global float* array) { array[get_global_id(0)] = 1; } with libclc, which would originally use the instrinsics like llvm.r600.read.local.size.x(). I executed the generated object file with one version of the hsa-runtime [1] provided by Mr. Stellard, when there was more than one workgroup, the output

Hi Mr. Liu, Thanks for your quick reply. I compiled the code with the libclc_trunk and linked the bitcode file under $LIBCLC_DIR/built_libs/tahiti-amdgcn--.bc. After looking into the libclc, it is currently using the new workitem intrinsics (commit ba9858caa1e927a6fcc601e3466faa693835db5e). In the linked bitcode ($LIBCLC_DIR/built_libs/tahiti-amdgcn--.bc), it has the following code segment,

Hello, I'd like to see support in clang/LLVM for multi-core parallelism, especially support for OpenMP. I think that the best way to do this is by designing an LLVM-based API (metadata and intrinsics) for expressing parallelism constructs, and having clang lower OpenMP code to that API. This will allow maximal preservation of optimization capabilities including target-specific lowering. What

Hi all, I would like my optimization pass to change an object's address space that is created by llvm.lifetime.start intrinsic. Because I want to be able to identify them later in a codegen pass. I can get a pointer from the intrinsic using CallInst::getArgOperand() function. However, I don't know what to do with it (or if it is the pointer that I want). How can I change its address space?

On Fri, Mar 25, 2016 at 02:22:11PM -0400, Jan Vesely wrote: > Hi Tom, Matt, > > I'm working on a project that needs few coherent atomic operations (HSA > mode: load, store, compare-and-swap) for std::atomic_uint in HCC. > > the attached patch implements atomic compare and swap for SI+ > (untested). I tried to stay within what was available, but there are > few issues

netbook (http://www.laptopspark.com) The world's largest computer maker (http://www.laptopspark.com/products/HP-G62-340US-NoteBook-AMD-Athlon-II-Dual-Core-P340220GHz-156-3GB-Memory-320GB-HDD-5400rpm-DVD-Super-Multi-ATI-Radeon-HD-4250-laptop-2047.html) startles analysts by scaling back its revenue forecast for the second time in as many quarters. It cites weak demand for desktops and

Hi, all I have a large number of NetCDF files that I need to average and then put the mean in a new NetCDF file. My problem is that I do this every time for files with different variables, so I do not want to hard-code the variables in the new NetCDF file I creat. does anyone know how I can copy ALL the dimensions and variables from one NetCDF file to another, and just put the mean I calculated as

I am unable to calculate confidence intervals for the slope estimate in a quasibinomial glm using confint(). Below is the output and the package info for MASS. Thanks in advance! R 2.9.2 MASS 7.2-48 > confint(glm.palive.0.str) Waiting for profiling to be done... Error: dims [product 37] do not match the length of object [74] > glm.palive.0.str Call: glm(formula = cbind(alive, red) ~ str,

Greetings, I am a very novice user with R, and in the course of running a linking procedure : P.old<- function (a, c, b, xi){ #a contains a parameters #c contains c parameters #b contains b parameters #xi is a one column vector containing quadrature points for xi(=theta-gamma)for one item tmp <- a*(xi-b)

Hey everyone, I have a series of thumbnails that have a loading overlay placed over them when they''re clicked on. The overlay is set to an opacity of .7.. .The onclick code looks basically does this: var loading = document.createElement(''div''); loading.id = ''loading_image''; $(loading).addClassName(''thumb_loading'');

Hi to all! I have a problem with a nvidia geforce 520mx [NVd0 generation card (0x0d9110a1)] i have on my notebook (samsung 3 series). In practice i'm not able to use it with bumblebee and bbswitch. The dmesg message is: [ 13.507435] nouveau 0000:01:00.0: power state changed by ACPI to D0 [ 13.507440] nouveau 0000:01:00.0: power state changed by ACPI to D0 [ 13.507448] nouveau

I'm looking for a sanity check on extending SLP Vectorizer to deal with aggregates. I'd like to vectorize Julia tuple operations. The Julia compiler lowers tuples to LLVM arrays, not LLVM vectors. I've tried making Julia lower tuples to LLVM vectors, but that hurt performance when SLP Vectorizer was not applicable, because of extraction/insertion overhead. I.e., the Julia lowering

Tom, I think it might be productive to fork this thread to discuss making the requirements for upstreaming a new LLVM target more explicit and open. I'd also like to gauge interest in an idea I've discussed privately with a few community members, namely the concept of having a semi-official "incubation" system whereby proposed backends could get a trial run before becoming part