search for: trigonometri

----- Original Message ----- > From: "Dimitri Tcaciuc" <dtcaciuc at gmail.com> > To: llvmdev at cs.uiuc.edu > Sent: Sunday, January 27, 2013 3:42:42 AM > Subject: [LLVMdev] SIMD trigonometry/logarithms? > > > > Hi everyone, > > > I was looking at loop vectorizer code and wondered if there was any > current or planned effort to introduce SIMD

----- Original Message ----- > From: "Elior Malul" <elior.malul at intel.com> > To: "Michael Gottesman" <mgottesman at apple.com>, "Hal Finkel" <hfinkel at anl.gov> > Cc: "LLVM Developers Mailing List" <llvmdev at cs.uiuc.edu> > Sent: Thursday, February 14, 2013 8:33:42 AM > Subject: RE: [LLVMdev] SIMD

Hi Justin, I think having .bc math libraries for different backends makes perfect sense! For example, in case of NVPTX backend we have the following problem: many math functions that are only available as CUDA C++ headers could not be easily used in, for instance, GPU program written in Fortran. On our end we are currently doing exactly what you proposed: generating math.bc module and then link

> From: "Dimitri Tcaciuc" <[hidden email]> > Sent: Sunday, January 27, 2013 3:42:42 AM > Subject: [LLVMdev] SIMD trigonometry/logarithms? > > I was looking at loop vectorizer code and wondered if there was any > current or planned effort to introduce SIMD implementations of > sin/cos/exp/log intrinsics (in particular for x86-64 backend)? I am currently

FWIW, llvmpipe has C-binding code that generates (inlines) log/exp/pow/sin/cos functions of the fly, for an arbitrary number of elements (in most cases), and uses sse/avx intrinsics as available: http://cgit.freedesktop.org/mesa/mesa/tree/src/gallium/auxiliary/gallivm/lp_bld_arit.c The precision is 20bits, which is enough for 3D. But it wouldn't be difficult to have more (or even variable)

Hi all. In fact, this is how we have implemented it in our compiler (intel's OpenCL). We have created a .bc file for every architecture. Each file contains all the SIMD versions for the functions to be vectorized. To cope with the massive amount of code to be produced, we implemented a dedicated tblgen BE for that purpose. We are willing to share that code with the llvm community, in case this

First let me say that I really like the notion of being able to plug in .bc libraries into the compiler and I think that there are many potential uses (i.e. vector saturation operations and the like). But even so it is important to realize the limitations of this approach. Generally implementations of transcendental functions require platform specific optimizations to get the best performance and

----- Original Message ----- > From: "Dmitry Mikushin" <dmitry at kernelgen.org> > To: "Justin Holewinski" <justin.holewinski at gmail.com> > Cc: "Hal Finkel" <hfinkel at anl.gov>, "LLVM Developers Mailing List" <llvmdev at cs.uiuc.edu> > Sent: Sunday, January 27, 2013 10:19:42 AM > Subject: Re: [LLVMdev] SIMD

Hi everyone, I was looking at loop vectorizer code and wondered if there was any current or planned effort to introduce SIMD implementations of sin/cos/exp/log intrinsics (in particular for x86-64 backend)? Cheers, Dimitri. -------------- next part -------------- An HTML attachment was scrubbed... URL:

I'm wondering if it makes sense to instead supply a bc math library. I would think it would be easier to maintain and debug, and should still give you all of the benefits. You could just link with it early in the optimization pipeline to ensure inlining. This may also make it easier to maintain SIMD functions for multiple backends. On Sun, Jan 27, 2013 at 8:49 AM, Hal Finkel <hfinkel

On 28/01/2013 12:49 AM, Hal Finkel wrote: > Ralf Karrenberg had implemented some of these as part of his > whole-function vectorization project: > https://github.com/karrenberg/wfv/blob/master/src/utils/nativeSSEMathFunctions.hpp > https://github.com/karrenberg/wfv/blob/master/src/utils/nativeAVXMathFunctions.hpp > Opinions on pulling these into the X86 backend? -Hal I've

...or non-finite inputs and outputs, the library should return the same results as libm. The library is tested if the evaluation error is within the designed limit. The library is tested against high-precision evaluation using the libmpfr library. Especially, we rigorously checked the error of the trigonometric functions when the arguments are close to an integral multiple of PI/2. The size of the functions is very small. Implementation of the library: Basically, each function consists of reduction and kernel. For the kernel, a polynomial approximation is used. The coefficients are carefully set to...

This may be a trivial question for some of you: Is there a way to add defined grid sizes on specific areas of a map?. I want to add 10kmx10km grids to all California coastal counties: "california,alameda", "california,contra costa", "california,del norte", "california,humboldt", "california,los angeles", "california,marin",

Dear all I am new R user, and trying to learn more. I am doing linear regression analysis in R with my data. I am trying to find the way to calculate the slope value (coefficient of x) to degree of slope. Please give me idea on this. Thanking you in anticipation Warm regardMS _________________________________________________________________ [[alternative HTML version

I write about R every weekday at the Revolutions blog: http://blog.revolution-computing.com . In case you missed them, here are some articles from last month of particular interest to R users. http://bit.ly/16wIdo offered a sneak peek of the debugger we've since released for subscribers of REvolution R Enterprise on Windows: http://bit.ly/1uxU3w . http://bit.ly/sZLaR brought news of big

Now I just need some helpful hints from the community as to what I could use to display a hexagonal playing field of about 40x40 hex units. Trying to build an online strategy game with Ruby on Rails. Working out the methodology on paper and then scaffolding some. Any ideas which technologies I could possibly use ? other than just simple .gifs as icons and maybe overlayed css graphics ? Would

The Vorbis audio format, uses something called the Modified Discrete Cosine Transform (MDCT). I was interested in how this worked, but did not know where to start. Is this the name of a particular algorithm, or does it just mean something like a cosine transform, but not quite right? So I asked the Vorbis mailing list. Nobody answered. It was not a high priority, so I did other things for

I'm having trouble understanding the output from as.windrose(). For one thing, data on a boundary between sectors seem to be left out of the counts. I assume that explains the missing point in the output below (angle 45). Shouldn't one side of each sector interval be open, to include values such as my 45 in the example? Also, why does the angle 180 in my input apparently not result in