howdy all! Just wanted to let y'all know that we have an #oneAPI meetup with David Airlie titled
"Why compute needs commoditization and currently sucks?"
Come join us this Friday at 4:30pm PT!
https://www.meetup.com/oneapi-community-us/events/295639963/
For those who have waited eagerly for the recording of our #oneapi meetup with @karolherbst - here it is! https://youtu.be/KUze0JbPSy8
#sycl #opencl #rustlang - if you are interested in joining our oneAPI meetup - feel free to subscribe here - https://www.meetup.com/oneapi-community-us/
Next time we will be meeting with Stephano Cetola who will be talking about RISC-V, onAPI, and other things.
#oneapi #sycl #opencl #rustlang
Next week #oneAPI Meetup will feature @karolherbst - they'll be talking about rusticl, how openCL works in Mesa, and varriers to #SYCL and any other questions people want to entertain. :)
https://www.meetup.com/oneapi-community-us/events/294379731/?isFirstPublish=true
Please boost to those who might be interested ! :)
(Thread)
Wanted to give a quick thanks to @sri for having me on as part of the #OneAPI Accelerated compute panel yesterday
Had a good time discussing what does and doesn't work across oneAPI, #SYCL and how to deal with SYCL as a standard.
I also want to reiterate that SYCL or oneAPI as standards are not "magic bullets" for performance across all hardware
The major point of them, like most libraries and software stacks, is to get you to ~90% of performance without having to rewrite the stack
Intel's Codeplay Announces oneAPI Construction Kit For Bringing SYCL To New Hardware
https://www.phoronix.com/news/oneAPI-Construction-Kit
"This open-source project aims to help ease bringing up SYCL on new processor/accelerator architectures, particularly around HPC and AI. The oneAPI Construction Kit also has a reference implementation for RISC-V."
#IamIntel #Intel #oneAPI #HeterogeneousComputing #GPU #GPGPU #HPC #AI #SYCL #RISCV
#iamintel #intel #oneapi #HeterogeneousComputing #gpu #gpgpu #hpc #ai #sycl #riscv
Intel Embree 4.1 is now available with real-time #rendering through #raytracing hardware acceleration on Intel GPUs — gain productivity using a single #SYCL codebase for both CPU and GPU rendering. Learn more: https://intel.ly/3Mtt7c9
Corporate #FLOSS at its worst: #NVIDIA controls the #Thrust library and its #CUDA, #OpenMP and #TBB backend. #AMD provides rocThrust, that is just Thrust with the CUDA part stripped an a new backend for #ROCm / #HIP. Nobody* is working on a backend for #SYCL
#Intel provides its own #oneAPI alternative as #oneDPL, which is NOT a drop-in replacement.
This is why we can't have nice things.
*there's a dead project here
https://github.com/wdmapp/syclthrust
#onedpl #oneAPI #intel #sycl #hip #ROCm #amd #tbb #OpenMP #CUDA #thrust #nvidia #floss
When we work on #SYCL, we're adding new capabilities to C++ to enable offloading to accelerators like GPUs and get huge performance. But what about mainstream C++ programmers? We're working on bringing these features to ISO C++ so every C++ developer will be able to exploit the performance of accelerators: https://www.youtube.com/watch?v=hqhs72ItT6s
Just announced at IWOCL / SYCLcon, the Khronos Group has released SYCL 2020 Revision 7.
See what changes were made: https://www.khronos.org/news/permalink/khronos-group-releases-sycl-2020-revision-7
#SYCL
Please to announce the CfP for the #oneAPI DevSummit on June 13th 2023 Online is open! Looking forward to seeing your submitted talks!
OK but for real: Since the name DPCPP is being deprecated in favor of ICPX with the -fsycl flag, can we please move to the name iSYCL?
Upstream name of the DPCPP project is staying as the reference implementation.
The intel packaged version no longer has a catchy name 😔
#isycl #intel #sycl #DPCPP
In slight need of spec/language lawyers: Reading this line ( https://registry.khronos.org/OpenCL/specs/3.0-unified/html/OpenCL_API.html#CL_DEVICE_MAX_MEM_ALLOC_SIZE), do you think:
A: that the minimum allocation of the max mem allocation flag is 1:4?
B: that the max allocation must be max(min(1024 × 1024 × 1024, 1/4th of CL_DEVICE_GLOBAL_MEM_SIZE), 32 × 1024 × 1024)
in other words, given a 16GB card, should the maximum allocation be 4GB, or should the minimum of the max allocation be 4GB?
#opencl #sycl #khronos #hpc #gpgpu
Introducing Khronos Discord! A new way for the community to interact with each other and learn more about Khronos related APIs (glTF, OpenXR, OpenCL, SYCL, etc...)
Help us build the community, join today!
https://www.khr.io/khrdiscord
#glTF #OpenXR #OpenCL #API #SYCL #OpenGL
#gltf #openxr #opencl #api #sycl #opengl
Why am I so interested in #SYCL for #GPUSPH? (See also my nudge nudge wink wink at @sri <https://fediscience.org/@giuseppebilotta/109942222462252885>)
Because SYCL is today the best bet we have at a unified #GPGPU API, and introducing such a backend would have great potential for our aims of “universal hardware support”.
The most important is definitely our dependency on #Thrust in a couple of places, but the extensive use of constant and shared memory is also a problem, since the syntax for their use in #SYCL is very different from the one in #CUDA, which would pose a problem for our plans to keep the device source code unified across APIs.
Here's another interesting thing about #GPUSPH on #Android: currently, it uses the #CPU backend, because the only #GPU backends available in #GPUSPH are for #NVIDIA #CUDA and #AMD through #HIP. I've actually looked at adding #SYCL support too, but there are a few structural issues that currently prevent a solution as straightforward as the one used to implement CPU and HIP support.
#sycl #hip #amd #CUDA #nvidia #gpu #cpu #android #GPUSPH
The meetup this week is going to rock - we have @illuhad giving a talk, but we also have Ronan from Tri-SYCL also showing up - mini #sycl con, ya'll? :-)
https://www.meetup.com/oneapi-community-us/events/292080220/
I wrote the (maybe final?) post on modern software development tools for oneAPI on Linux -
https://dev.to/oneapi/modern-software-development-tools-and-oneapi-part-3-3dok
#sycl #oneapi #gnome #linuxapps
#sycl #oneapi #gnome #linuxapps