search for: __immediate

...he use cases at hand and design specific analysis > (and also transformation) passes "from scratch". The long term goal > should be a full scheduling pipeline but the parts need to be designed > in a modular (LLVM-way) from the very beginning. This allows us to: > - provide __immediate benefit__ to other developers, > - allow active participation in (and understanding of) the design of > each part, and > - develop the intermediate parts with the requirements of the whole > LLVM project in mind. > > Let me give two examples to make my point: >...

...hand and design specific analysis > >(and also transformation) passes "from scratch". The long term goal > >should be a full scheduling pipeline but the parts need to be designed > >in a modular (LLVM-way) from the very beginning. This allows us to: > > - provide __immediate benefit__ to other developers, > > - allow active participation in (and understanding of) the design of > > each part, and > > - develop the intermediate parts with the requirements of the whole > > LLVM project in mind. > > > >Let me give two example...

...e should start with the use cases at hand and design specific analysis (and also transformation) passes "from scratch". The long term goal should be a full scheduling pipeline but the parts need to be designed in a modular (LLVM-way) from the very beginning. This allows us to: - provide __immediate benefit__ to other developers, - allow active participation in (and understanding of) the design of each part, and - develop the intermediate parts with the requirements of the whole LLVM project in mind. Let me give two examples to make my point: -- Example 1 -- In addition to the ap...

...uld start with the use cases at hand and design specific analysis (and also transformation) passes "from scratch". The long term goal should be a full scheduling pipeline but the parts need to be designed in a modular (LLVM-way) from the very beginning. This allows us to: > - provide __immediate benefit__ to other developers, > - allow active participation in (and understanding of) the design of > each part, and > - develop the intermediate parts with the requirements of the whole > LLVM project in mind. > > Let me give two examples to make my point: > &g...

...ific analysis > > >(and also transformation) passes "from scratch". The long term goal > > >should be a full scheduling pipeline but the parts need to be designed > > >in a modular (LLVM-way) from the very beginning. This allows us to: > > > - provide __immediate benefit__ to other developers, > > > - allow active participation in (and understanding of) the design of > > > each part, and > > > - develop the intermediate parts with the requirements of the whole > > > LLVM project in mind. > > > > &g...

** *Hi everyone,As you may know, stock LLVM does not provide the kind of advanced loop transformations necessary to provide good performance on many applications. LLVM's Polly project provides many of the required capabilities, including loop transformations such as fission, fusion, skewing, blocking/tiling, and interchange, all powered by state-of-the-art dependence analysis. Polly also