Customized Parallel Computing research group retweeted
Our article introduces a dual-mode processor that utilizes ILP statically when available, without suffering from low code density when it's lacking. The architecture, supported by a novel compilation method, enables fine-grained mode switching. Read more: ieeexplore.ieee.org/document…
Customized Parallel Computing research group retweeted
PoCL-R is a new backend for offloading OpenCL tasks to other nodes on the network. Now with OpenCL API offloading can be performed identically whether using local or remote devices - very interesting for adaptive edge offloading and other use cases! portablecl.org/remote-backen…
Customized Parallel Computing research group retweeted
VLIW processors exploit ILP efficiently but suffer from bad code density in serial parts where instruction packets cannot be fully utilized. Read how we increased flexibility and energy efficiency by using both a TTA and RISC-V based ISA in one processor.
researchgate.net/publication…
Customized Parallel Computing research group retweeted
The open-standard RISC-V ISA has made the open source hardware community more active but there has been a lack of tools for customized RISC-V processors. Read how we added support for RISC-V based ASIPs to the upcoming 2.0 release of the OpenASIP toolset.
researchgate.net/publication…
Customized Parallel Computing research group retweeted
Emerging memory technologies could enable extremely energy-efficient compute devices in the future. Read how domain wall memories can be utilized efficiently with our SHRIMP method.
ieeexplore.ieee.org/abstract…
@TampereUni @CustomParComp @cfaed_TUD @FitOptiVis @CPSoSAWARE
Customized Parallel Computing research group retweeted
Did you know that PoCL is also an excellent OpenCL application debugging aid? We added more documentation for this use case: portablecl.org/docs/html/deb… @openclapi @CustomParComp
Customized Parallel Computing research group retweeted
Read how our SHRIMP method developed in collaboration with @cfaed_TUD
helps in reducing costly shifts, when executing code from racetrack memory.
researchgate.net/publication…
@TampereUni @CustomParComp @SuomenAkatemia @FitOptiVis @BusinessFinland
Customized Parallel Computing research group retweeted
POCL 1.5 released with support for #LLVM 10, an easy to use kernel profiling features and plenty of fixes and performance improvements: portablecl.org/pocl-1.5.html @openclapi @thekhronosgroup @FitOptiVis @HSAFOUNDATION @ComputingSc_TAU
Customized Parallel Computing research group retweeted
TCE 1.21 released! It got #LLVM 10 support, spill code alias analyzer, a simplified minimal load-store unit... openasip.org/download.html @ComputingSc_TAU #openhw @FitOptiVis @HSAFOUNDATION @llvmorg @CustomParComp
Customized Parallel Computing research group retweeted
Provost Jarmo Takala and Professor Atanas Gotchev @TampereUni welcome you to Tampere in September 2020 for the #MMSP2020 - the IEEE 22nd International Workshop on Multimedia Signal Processing! youtu.be/YzVBHjbeeY0 @VisitTampere #MMSP #IEEE #SignalProcessingSociety
Customized Parallel Computing research group retweeted
Let's together make 1.4 the best pocl release yet! github.com/pocl/pocl/wiki/Re…
Customized Parallel Computing research group retweeted
Our real-time path tracing denoising paper BMFR will be presented next week at #SIGGRAPH2019 See you all in room 153 on Jul 30 10:45 AM PDT tuni.fi/vga/bmfr
Customized Parallel Computing research group retweeted
POCL received support for CL_DEVICE_TYPE_CUSTOM via contribution of a hardware accelerator framework! For more information: computer.org/publications/te… portablecl.org/docs/html/acc… @HSAFOUNDATION @openclapi @ITC_TampereUni @CustomParComp @FitOptiVis @thekhronosgroup
Customized Parallel Computing research group retweeted
TCE 1.19 with LLVM 8.0 support and usability improvements released! openasip.org/download.html
Customized Parallel Computing research group retweeted
pocl 1.3 released! portablecl.org/pocl-1.3.html
Customized Parallel Computing research group retweeted
Time to test a new pocl release! Please help us shaking any remaining regressions out by testing and reporting your results as instructed here:
github.com/pocl/pocl/wiki/Re…
Customized Parallel Computing research group retweeted
pocl is powering @ThinkSilicon ultra-low power GPGPUs! portablecl.org/pocl-in-think… @CustomParComp @ComputingSc_TAU @FitOptiVis
Customized Parallel Computing research group retweeted
@teknsaatio made a blog post about our doctoral student @Koskela_Matias tekniikanedistamissaatio.fi/…
Customized Parallel Computing research group retweeted
It will be an exciting week for our group since, we will present a ray tracing hardware paper at HPG'18 this weekend (preprint now up at tut.fi/vga/publications/PLOC…), and also a ToG paper at SIGGRAPH in the beginning of the next week (tut.fi/vga/publications/Merg…).
Customized Parallel Computing research group retweeted
New #LLVM release branched, thus time for a new pocl release also! Please help testing the upcoming 1.2 with your favourite #OpenCL app: github.com/pocl/pocl/wiki/Re… @CustomParComp @PervasiveTUT @jrprice89