search for: raytrac

I've recently played with the GCC implementation of pointer checker on a real hardware, my recent impressions are here: https://github.com/google/sanitizers/wiki/AddressSanitizerIntelMemoryProtectionExtensions (there is also some old pre-hardware content). In short, I totally agree with what David says above: MPX is a disaster. (Usual disclaimer: my opinion here is too biased) I am glad

...rsion is 5.2.1, icc version 2016.1.150. The > benchmark suite is PARSEC 3.0, all versions with 1 thread and default > configs. > > As I described previously, PointerChecker in gcc produces very > inefficient code. My experiments show overheads over native of up to > 9.5X (on "raytrace"), with common overheads of 3X ("bodytrack", > "fluidanimate", "streamcluster"). At the same time, AddressSanitizer > performs much better -- 1.3X on "raytrace", 1.7X on "bodytrack" and so > on. > > Recently I played with MPX su...

...us:~/m/autotune|master⚡*? ➤ /Users/kavon/msr/llvm5/bin/opt -inline -mem2reg -inferattrs -ipconstprop -gvn -simplifycfg -bdce -sink -dse -adce -instcombine -early-cse-memssa -early-cse-memssa -dse -adce -simplifycfg -sink -ipconstprop -gvn -bdce -instcombine -jump-threading -inline -sroa ./src/apps/raytracer.bc -o ./out/raytracer_opt1043.bc 0 opt 0x000000010ebd4498 llvm::sys::PrintStackTrace(llvm::raw_ostream&) + 40 1 opt 0x000000010ebd4a46 SignalHandler(int) + 342 2 libsystem_platform.dylib 0x00007fff8ce5352a _sigtramp + 26 3 opt...

...Thu, Feb 4, 2016 at 4:59 AM, Dmitrii Kuvaiskii < Dmitrii.Kuvaiskii at tu-dresden.de> wrote: > >> Recently I played with MPX support on Intel C/C++ Compiler (icc). This > >> implementation looks *much* better, with the following example > >> overheads: 1.2X on "raytrace", 1.25X on "bodytrack", 1.08X on > >> "streamcluster". So the common overheads are in the range of 15%-25%! > > That's interesting. > > Are you sure you are instrumenting both reads and writes with icc? > > Yes, here are the exact flags I add...

...cc based D compiler) and g++. I used a ray tracing program found in this thread (make sure to use the code mentioned later in the thread as it has some optimizations): Go to digitalmars.com/ webnews/ newsgroups.php?renew=1 and search for: "D slower than C++ by a factor of _two_ for simple raytracer (gdc)" There are versions of the tracer for D and C++ that are quite comparable. Here are some timings I got on an AMDx86-64 running Fedora Core Linux. Timings are averaged over 6 runs. llvm-g++4.0.1 5.76 (build 5449..backend from svn today) ldc-rev736 6.68 g++4.1.2 6...

...nd g++. > I used a ray tracing program found in this thread (make sure to use > the code mentioned later in the thread as it has some optimizations): > > Go to digitalmars.com/ webnews/ newsgroups.php?renew=1 and search > for: “D slower than C++ by a factor of _two_ for simple raytracer > (gdc)” Wow, very very nice! Have you pointed this out to the other D users? -Chris > > There are versions of the tracer for D and C++ that are quite > comparable. Here are some timings I got on an AMDx86-64 running > Fedora Core Linux. Timings are averaged over 6 runs....

...iskii < >> Dmitrii.Kuvaiskii at tu-dresden.de> wrote: >> >>> >> Recently I played with MPX support on Intel C/C++ Compiler (icc). This >>> >> implementation looks *much* better, with the following example >>> >> overheads: 1.2X on "raytrace", 1.25X on "bodytrack", 1.08X on >>> >> "streamcluster". So the common overheads are in the range of 15%-25%! >>> > That's interesting. >>> > Are you sure you are instrumenting both reads and writes with icc? >>> >>...

...more precise term) specialised for graphics. I hope to allow domain specific optimizations, eg folding together post-processing passes, or automatic LODing for procedural textures. I'm planning to make the renderer itself to be defined at runtime, to allow foward rendering, deferred rendering, raytracing, or a combination of all 3. Of course, this is lot to chew on, so for now it's more of a proof of concept than anything production-ready. Due to my unfamiliarity with LLVM, I do have some broad questions though. If somebody can answer, or just point me to the relevant documentation/code, I...

Hi Mike, On 2013-08-02 18:26, Michael Lewis wrote: > I've been working recently on a precise garbage collector which runs > alongside native code JITted by LLVM. Today marks the first time the > GC has passed its entire test suite as well as extensive soak tests in > non-trivial programs. I'm glad to hear that you solved the issues you described in your last mail. How did you

...>//>>>>>>>>/Recently I played with MPX support on Intel C/C++ Compiler (icc). />>>>>>>>/This />>>>>>>>/implementation looks *much* better, with the following example />>>>>>>>/overheads: 1.2X on "raytrace", 1.25X on "bodytrack", 1.08X on />>>>>>>>/"streamcluster". So the common overheads are in the range of />>/15%-25%! />>>>>>>/That's interesting. />>>>>>>/Are you sure you are instrumenting both re...

Hello, I set up an script witch connects my rsyncd server. But the IPs for rsyncd Server and Client have no DNS Name, and so rsync trys to lookup for a DNS Name for this IPs. This causes a long DNS Timeout! My question: is there a commandline option to disable this "feature"? Or can you give me quick "hack around"? Best regards -- Clemens Gesell <cgesell@astaro.de> |

Hi all, I've been working recently on a precise garbage collector which runs alongside native code JITted by LLVM. Today marks the first time the GC has passed its entire test suite as well as extensive soak tests in non-trivial programs. It's been an interesting and educational process, to say the least, and I've run into quite a few things that might be useful to know for others

I have a problem with function 'inout()' in package 'splancs' on CRAN-R Version 1.2.3 under FreeBSD4.3-STABLE. The following script produces and draws points and a polygon-surrounded area. Repeating the same script many times shows, that points on the polygon-line often, but not always, are outside of the polygon-area. library(splancs) # dataset with polygon (convex

...calculating the surface normals I do now, with a Sobel filter or perhaps something more advanced). 3) Triangulate the first slice/frame 4) Using a cross product between the edge in the first slice and the gradient normal, calculate a second vector to make the edges into planes. Find their edges by raytracing, like in the 2D version. Alternatively, use some other algorithm (Marching Cubes springs to mind, but there are patents surrounding it) to extract planes in the data. I'm not really sure whether this would work. It doesn't seem very logical to me for some reason. Also I don't know...

Hi, I guess this is a good time to start putting together a wish list for a video codec. I see that for audio the compression is around 10X for reasonable quality. I am sure this will start its own thread of conversation. For video you can do 40X fairly easily and the big task is to go to 80X or 100X with reasonable picture quality, say, a peak luma SNR of more than 30 dB. Uncompressed