Hello, On Wed, Jan 21, 2015 at 11:14 AM, Krishnaraj Raghavendra Bhat <k.r.b...@samsung.com> wrote: > We would like to propose openmp optimizations in some parts of evas software > backend > > OpenMP provides a scalable model for shared memory parallel applications. > One of the construct enables parallelization of simple for-loops using > 'pragma omp for'. We applied this pragma in one part of sample scaling in > software backend where for-loop was used to run through the image. Running > expedite on an Intel I3 machine with quad-core showed 48% improvement in > "Image Blend Nearest Scaled" test-case. Test-case score went up from 128 to > 190
Could you provide before and after full result from expedite. We had in the past some parallelisation code and it ended up triggering slower path for most case. It also resulted in consuming massively more CPU with little improvement for the few tests case where we could see an improvement. Basically, if by adding 1 core, you gain 30% and now both core are running at 100%, we are not sure it is a good tradeof. But if by having 3 more cores, you only get 48% speed increase and all core are running at 100%, then we are sure it is not a wise choice from an energy perspective. > This doesn't mean that every for-loop will be benefited by openmp. For-loop > should have enough parallel-load and no dependencies so that threading > overhead is hidden. Other areas of data-parallelism in evas can be analyzed > and explored further. Openmp is supported in gcc since 4.2. Pragma based omp > programs require a light-weight gcc omp runtime(libgomp.so) which is always > packaged with gcc-runtime We need extensive benchmark on speed, memory and battery consumption for this move. One of the change we can investigate is to do that manually and tweak it to only start a meaningful number of core depending on the task. In many case the main issue is memory bandwidth and we can maybe have some gain first by doing light compression on some more data. Anyway, it's an area that require a lot of experimentation and data to move forward. We are definitively interested by more information. -- Cedric BAIL ------------------------------------------------------------------------------ New Year. New Location. New Benefits. New Data Center in Ashburn, VA. GigeNET is offering a free month of service with a new server in Ashburn. Choose from 2 high performing configs, both with 100TB of bandwidth. Higher redundancy.Lower latency.Increased capacity.Completely compliant. http://p.sf.net/sfu/gigenet _______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
