| Table of Contents | ||
|---|---|---|
|
Attendees
Larry - Cavium
Kangkang - Huawei
Tianghong - Huawei
Thomas - ARM
Sean - Qualcomm
Takeharu - Fujitsu
Scott Hare - Qualcomm
David Brash - ARM
Renato - Linaro
Gema Gomez - Linaro
Graeme G - Linaro
Ganesh Raju - Linaro
Anoop S - Linaro
Kanta Vekaria- LInaro
Mike Holmes - Linaro
Francois - Linaro
Martin - Linaro
Agenda
- Technical discussion on Tooling
View file name HPC TCWG BUD17.pdf height 250
Minutes
Kanta : Started discussion with HPC SIG SC Engineering Prioritization discussion
Waiting for ARM to release spec before we can schedule tasks for optimising SVE.
Renato: Only bullet point is “review SVE patches”. There is lot to be done before even looking at SVE
Larry: Enough low hanging fruit in HPC now SVE is not a priority
Kanta: how important is SVE for QEMU?
Fujitsu: Important for us.
Larry: how do we compliment ARM Manchester HPC team?
Tianghong: Enabling in OpenHPC a priority!
Kanta: Linaro can do initial analysis
Quest: Do we have a plan for multi arch aarch64 microarchitecture support (ie ThunderX vs other Hisilcon)
Renato: asked if this is per system, per cluster or other deciding point. Need to chose how this is transparent for OpenHPC.
Kanta: Key focus area is OpenHPC
From Linaro next step is to investigate what is in OpenHPC, libraries available
Build list of libraries
TCWG
Started presentation for TCWG plans
Initial focusing on LLVM Compiler for (Loop optimization & OpenMP)
FASTMath is not the same as in LLVM vs GCC
We need some kind of benchmarks for HPC
Renatto: Doing single and double precision,
TCWG: not doing half precision. We need SIMD enabling for half precision, but should not matter much.
Talking to Intel on what is missing in OpenMP and talking to ARM about performance
Libomp Missing component for OpenMP 4.0 is Offloading, To be finalize release cycle
Starting to make sure all the libraries are available into OpenHPC (libc++, libomp, etc)
Larry: Is there any defined process for OpenHPC. Need Eric’s help
Kanta: Not currently and working with their Steering committee
With SVE will collaborate with ARM and upstream libraries. Make sure they are as much complete for LLVM. And it will take approx 3 months
Looking at SVE support in GDB.
GCC engineers are also working LLVM and optimizations are made in both at the same time.
Resource: With past experience, proposal is that, to work with member engineers and technically lead them
Tianghong: Did we consider OCR (open community runtime). Do we have a plan for “”open community runtime””, It’s for application, inclu OSN. And plan for benchmark (European Conference on Applications of Surface and Interface Analysis (ECASIA)).
Kanta: Maybe in the future we should look at the OCR efforts and compare with OpenHPC and see if it can be combined.
Benchmarking - UEABS
Action: Eric V(ARM) to highlight what benchmarks are in openHPC
TCWG: Benchmark shows libomp performances better than libgomp but libgomp is more compliance and libgomp comes with GCC. If the libraries are not available use libgomp and if performance is needed move to libomp.
Will start working on this, once received feedback from HPC, move all LLVM libraries to OpenHPC
Martin: Priority to start with Overlay, PPA
Renatto: Aboot ARM environment ?
Actions
Re-evaluate OpenHPC in terms of efforts
Benchmarking : Need expertise from the people who are doing this
Quest: Benchmarking on SIngle or multi node