Commits

Breadcrumbs

History for

ondecrease-verbosity-ccdb-api

User selector

All users

All time

Commit history

Commits on Jun 26, 2019

• Workaround for broken address space deduction by specification of clang OpenCL

  Show description for c7c06fe
  davidrohr

  committed
  c7c06feView commit details

  Copy full SHA for c7c06fe

  View code at this pointBrowse repository at this point

• Simplify code

  davidrohr

  committed
  f380e32View commit details

  Copy full SHA for f380e32

  View code at this pointBrowse repository at this point

• Reduce default number of GPU threads for HIP

  davidrohr

  committed
  fa36de2View commit details

  Copy full SHA for fa36de2

  View code at this pointBrowse repository at this point

• Revert workaround for OpenCL C++

  Show description for dfd1f5f
  davidrohr

  committed
  dfd1f5fView commit details

  Copy full SHA for dfd1f5f

  View code at this pointBrowse repository at this point

Commits on Jun 23, 2019

• Several fixes to buffer overflow handling in TPC tracking

  davidrohr

  committed
  5430c41View commit details

  Copy full SHA for 5430c41

  View code at this pointBrowse repository at this point

Commits on Jun 20, 2019

• Remove remaining AliTPCCommon install files, no longer needed

  davidrohr

  committed
  c136e86View commit details

  Copy full SHA for c136e86

  View code at this pointBrowse repository at this point

Commits on Jun 4, 2019

• Clean up some horrid #ifdef magic, and fix some more OpenCL-C++ compilation (+ fix AliRoot Compilation)

  davidrohr

  committed
  397f155View commit details

  Copy full SHA for 397f155

  View code at this pointBrowse repository at this point

Commits on May 30, 2019

• Revert workaround for missing __generic old-style atomic overloads in OpenCL CPP spec (Spec Issue 66 ; clang bug 41963 / review D62335)

  davidrohr

  committed
  8bb8a43View commit details

  Copy full SHA for 8bb8a43

  View code at this pointBrowse repository at this point

• Revert "Temporary fixes to workaround clang bug 41567" - fixed in clang 9 (review D62299)

  Show description for f5d2b5e
  davidrohr

  committed
  f5d2b5eView commit details

  Copy full SHA for f5d2b5e

  View code at this pointBrowse repository at this point

• Revert "Reenable constant memory cache for GPU HIP, the problem has been fixed in the meantime is this workaround is no longer needed" - problem is fixed, but current AMD GPUs simply do not have su…

  Show description for 0bf7152
  davidrohr

  committed
  0bf7152View commit details

  Copy full SHA for 0bf7152

  View code at this pointBrowse repository at this point

• Reenable constant memory cache for GPU HIP, the problem has been fixed in the meantime is this workaround is no longer needed

  davidrohr

  committed
  3ef972bView commit details

  Copy full SHA for 3ef972b

  View code at this pointBrowse repository at this point

• Unify TPC constants as far as possible

  davidrohr

  committed
  8ad14f0View commit details

  Copy full SHA for 8ad14f0

  View code at this pointBrowse repository at this point

Commits on May 22, 2019

• HIP supports __clz (as CUDA), so we should use it

  davidrohr

  committed
  2ba7cf1View commit details

  Copy full SHA for 2ba7cf1

  View code at this pointBrowse repository at this point

• Added GPU sort functions with comparison object and cooperatively in block (placeholder for now)

  davidrohr

  committed
  21e9f2fView commit details

  Copy full SHA for 21e9f2f

  View code at this pointBrowse repository at this point

• Implement truncation of charge and with of TPC clusters to n significant bits

  davidrohr

  committed
  232e9b0View commit details

  Copy full SHA for 232e9b0

  View code at this pointBrowse repository at this point

• Implement Framework for TPC Data Compression (CPU + GPU) (No track model / differences yet)

  davidrohr

  committed
  3ddca84View commit details

  Copy full SHA for 3ddca84

  View code at this pointBrowse repository at this point

Commits on May 18, 2019

• Do not pollute global namespace outside of AliGPU

  davidrohr

  committed
  08e0580View commit details

  Copy full SHA for 08e0580

  View code at this pointBrowse repository at this point

• Workaround for GPU Constant memory usage with HIP, to make symbol not optimized out (not yet working due to another bug in HIP)

  davidrohr

  committed
  0305c31View commit details

  Copy full SHA for 0305c31

  View code at this pointBrowse repository at this point

Commits on May 16, 2019

• Fix standalone build without O2 headers

  

  davidrohr

  authored and

  ktf

  committed
  87e41eeView commit details

  Copy full SHA for 87e41ee

  View code at this pointBrowse repository at this point

• Fix TYPO

  

  davidrohr

  authored and

  ktf

  committed
  d6690ebView commit details

  Copy full SHA for d6690eb

  View code at this pointBrowse repository at this point

Commits on May 10, 2019

• Implement GPU based TPC transformation

  

  davidrohr

  authored and

  shahor02

  committed
  0fa923cView commit details

  Copy full SHA for 0fa923c

  View code at this pointBrowse repository at this point

• Force CUDA stack size (nvcc cannot determine correct stack size due to recursive sorting function, yielding a stack overflow)

  

  davidrohr

  authored and

  sawenzel

  committed
  2603ffbView commit details

  Copy full SHA for 2603ffb

  View code at this pointBrowse repository at this point

Commits on May 2, 2019

• Fix GPUTPCGeometry to work on GPU, required for dEdx on GPU

  davidrohr

  committed
  1abe291View commit details

  Copy full SHA for 1abe291

  View code at this pointBrowse repository at this point

Commits on Apr 30, 2019

• Add missing headers for installation

  davidrohr

  committed
  3ced571View commit details

  Copy full SHA for 3ced571

  View code at this pointBrowse repository at this point

• Workaround for CTests after Dario rewrote the CMake, to be reversed after AliTPCCommon is removed from alidist

  davidrohr

  committed
  31974eaView commit details

  Copy full SHA for 31974ea

  View code at this pointBrowse repository at this point

• First GPU TPC dEdx implementation

  davidrohr

  committed
  8833319View commit details

  Copy full SHA for 8833319

  View code at this pointBrowse repository at this point

• Temporary fixes to workaround clang bug 41567

  davidrohr

  committed
  2fb0110View commit details

  Copy full SHA for 2fb0110

  View code at this pointBrowse repository at this point

• Fix copy&paste typo

  davidrohr

  committed
  62304bfView commit details

  Copy full SHA for 62304bf

  View code at this pointBrowse repository at this point

• With sorting on GPU working generally, we can include ITS and TRD to HIP/OpenCL compilation

  davidrohr

  committed
  a8ec67aView commit details

  Copy full SHA for a8ec67a

  View code at this pointBrowse repository at this point

• Make a couple of files compile with OpenCL (mask unsupported stuff / move from header to cxx)

  davidrohr

  committed
  ff29b2bView commit details

  Copy full SHA for ff29b2b

  View code at this pointBrowse repository at this point

• Add GPUCommonAlgorithm class provinding GPU based sorting (in thread only for now)

  davidrohr

  committed
  87961b3View commit details

  Copy full SHA for 87961b3

  View code at this pointBrowse repository at this point

• Adapt clang-format 8, fix coding style

  davidrohr

  committed
  f8d4423View commit details

  Copy full SHA for f8d4423

  View code at this pointBrowse repository at this point

• Fix shared memory casts for old-style OpenCL atomics

  davidrohr

  committed
  eb45301View commit details

  Copy full SHA for eb45301

  View code at this pointBrowse repository at this point

• So far, OpenCL cannot cast between __generic and __constant, so we need to avoid using __constant for now

  davidrohr

  committed
  0771498View commit details

  Copy full SHA for 0771498

  View code at this pointBrowse repository at this point

• Add setting to place processors either in constant memory or in global memory on GPU

  davidrohr

  committed
  7decb4bView commit details

  Copy full SHA for 7decb4b

  View code at this pointBrowse repository at this point