search for: polyhedra

...the LLVM backend) > * We see non-trivial speedups for hmmer, libquantum, and various > linear-algebra kernels (we use gemm-specific optimizations). The > first two require additional flags to be enabled. On the topic of performance, this paper might be of interest: Fast Polyhedral Abstract Domain Gagandeep Singh, Markus Püschel, Martin Vechev http://www.srl.inf.ethz.ch/papers/POPL17-Polyhedra.pdf > Our experimental results demonstrate massive gains in both space and > time: we show end-to-end speedups of two to five orders of magnitude > compared to state-of-the-...

...* We see non-trivial speedups for hmmer, libquantum, and various >> linear-algebra kernels (we use gemm-specific optimizations). The >> first two require additional flags to be enabled. > > On the topic of performance, this paper might be of interest: > > Fast Polyhedral Abstract Domain > Gagandeep Singh, Markus Püschel, Martin Vechev > http://www.srl.inf.ethz.ch/papers/POPL17-Polyhedra.pdf > >> Our experimental results demonstrate massive gains in both space and >> time: we show end-to-end speedups of two to five orders of magnitude >&gt...

...C++ bindings (_http://repo.or.cz/isl.git/shortlog_). While a first set of patches is already open, further patches will follow over the next couple of weeks. Let me briefly summarize the LLVM developer meeting comments on our proposal (subjective summary) * Most people were interested in having polyhedral loop optimizations being part of LLVM. * Ideas of uses of isl beyond polyhedral loop scheduling were raised (e.g., for polyhedral value analysis, dependence analysis, or broader assumption tracking). Others were interested in the use of polyhedral loop optimization with “learned” scheduling strat...

The volume of a polygon = 0. Polyhedra have volumes. This may be irrelevant, but if the lake is cylindrical == constant cross sectional area at all depths, then height doubles when the volume does and vice versa. Otherwise you have to know how area varies with height or use more sensible approximations thereto. Cheers, Bert On Tue,...

Hi all, I'm trying to use the PPL (Parma Polyhedra Library, bugseng.com/products/ppl) in one of my LLVM passes getting the error: > Error opening > '/home/jorge/SvnReps/Systems/llvm/build/Debug+Asserts/lib/MyAnalysis.so': > /home/jorge/SvnReps/Systems/llvm/build/Debug+Asserts/lib/MyAnalysis.so: > undefined symbol: _ZNK23Parma_P...

Hi Jorge, > I'm trying to use the PPL (Parma Polyhedra Library, > bugseng.com/products/ppl) in one of my LLVM passes getting the error: > >> Error opening >> '/home/jorge/SvnReps/Systems/llvm/build/Debug+Asserts/lib/MyAnalysis.so': >> /home/jorge/SvnReps/Systems/llvm/build/Debug+Asserts/lib/MyAnalysis.so: >> undefi...

On Tuesday, November 15, 2011 at 10:11:25 (+0100), Duncan Sands wrote: >> Hi Jorge, >> >> > I'm trying to use the PPL (Parma Polyhedra Library, >> > bugseng.com/products/ppl) in one of my LLVM passes getting the error: >> > >> >> Error opening >> >> '/home/jorge/SvnReps/Systems/llvm/build/Debug+Asserts/lib/MyAnalysis.so': >> >> /home/jorge/SvnReps/Systems/llvm/build/Deb...

Hello The excellent adapt package integrates over multi-dimensional hypercubes. I want to integrate over a multidimensional simplex. Has anyone implemented such a thing in R? I can transform an n-simplex to a hyperrectangle but the Jacobian is a rapidly-varying (and very lopsided) function and this is making adapt() slow. [ A \dfn{simplex} is an n-dimensional analogue of a triangle or

Dear all; I am trying to calculate the volume of a polygon shapefile according to a DEM raster. I have provided some codes at the end of this email.I dont know if the codes are correct or not. Following this, I have another question too. I want to know if the volume of the reservoir rises or doubles, what would be the elevation? I would be more than happy if anyone could help me. Sincerely "

...m scratch. For instance, in LLVM, modeling PHI nodes as in Swift's intermediate representation might be such an idea, but requires huge changes in the code base. As contributor to Polly, I am no exception. I therefore would like to share my thoughts about what the best possible integration of a polyhedral optimizer/Polly could look like, to discuss which end result we might strive for. I am a proponent of polyhedral compilation and would like to see it used on a daily basis by just choosing an optimization level. IMHO we should avoid multiple loop optimizations in LLVM. If each of them does its ow...

...ling signed division/remainder operations, which worked around lack of support in ScalarEvolution, is being replaced thanks to improvements being contributed to ScalarEvolution itself (see D34598). Polly’s core delinearization routines have long been a part of LLVM itself. * PolyhedralInfo, which exposes a subset of Polly’s loop analysis for use by other clients, is now available. * Polly is now part of the LLVM release process and is being included with LLVM by various packagers (e.g., Debian). I believe that the LLVM community would benefit from beginning the...