log src/gpu/ptx/vm/gpu_ptx.cpp @ 13922:0995dcbd6dd8

age author description
Mon, 10 Feb 2014 14:38:42 -0500 S.Bharadwaj Yadavalli Change CUDA context management to support multiple executions of a kernel. Exclude GPU offloading of lambdas from java.* library code.
Thu, 06 Feb 2014 11:14:19 +0100 Doug Simon pulled Ptx and Hsail classes out of gpu class namespace
Fri, 31 Jan 2014 16:24:38 +0100 Doug Simon fixed non-PRODUCT build compilation error
Fri, 31 Jan 2014 16:05:37 +0100 Doug Simon removed PTXKernelArguments; added mechanism for printing incoming kernel arguments
Fri, 31 Jan 2014 00:32:03 +0100 Doug Simon Merge.
Fri, 31 Jan 2014 00:13:12 +0100 Doug Simon disable printing of CUDA errors unless TraceGPUInteraction is enabled
Thu, 30 Jan 2014 23:52:34 +0100 Doug Simon implemented GC locking for duration of a PTX kernel call
Thu, 30 Jan 2014 17:49:56 -0500 S.Bharadwaj Yadavalli Add additional device checks and compute capability checks for CUDA devices found on Linux.
Thu, 30 Jan 2014 00:52:33 +0100 Doug Simon added support for co-existing GPU backends (JBS:GRAAL-1)
Fri, 24 Jan 2014 18:01:36 +0100 Doug Simon partially fixed passing of object parameters to PTX kernels; use a C++ object for managing resource allocation and cleanup around a PTX kernel execution
Fri, 17 Jan 2014 16:03:13 -0500 S.Bharadwaj Yadavalli schedule lambda method compilation and execution on GPU (PTX) when possible; fix a couple of bugs.
Thu, 16 Jan 2014 17:50:58 +0100 Andreas Woess remove unreferenced local variable in gpu_ptx.cpp
Wed, 15 Jan 2014 20:24:13 +0100 Doug Simon added support for @ParallelOver annotation in PTX kernel call wrapper
Mon, 13 Jan 2014 22:32:27 +0100 Doug Simon initial code for calling PTX kernel code from Java with parameter marshaling and return value unmarshaling performed by a wrapper specified via manual graph construction
Fri, 01 Nov 2013 18:34:03 -0400 S.Bharadwaj Yadavalli * PTX kernel argument buffer now has naturally aligned arguments as required by PTX JIT compiler.
Tue, 22 Oct 2013 14:35:41 -0400 S.Bharadwaj Yadavalli PTX codegen enhancements; fixes to PTX test regressions.
Sun, 06 Oct 2013 22:07:23 -0400 Morris Meyer PTX initial float and double
Sun, 06 Oct 2013 18:15:56 -0400 Morris Meyer PTX boolean return value, emitIntegerTestMove, warnings
Sat, 05 Oct 2013 10:31:18 -0400 Morris Meyer Fixed BasicPTXTest and IntegerPTXTest
Mon, 30 Sep 2013 21:46:41 +0200 Andreas Woess gtx_ptx.cpp: fix switch default label and operator precedence in ncores().
Mon, 30 Sep 2013 13:03:47 -0400 Morris Meyer PTX warp limiter to available GPU processors
Sun, 29 Sep 2013 14:47:12 -0400 Morris Meyer PTX: one-dimensional parallel warp invocation, ParallelOver annotation
Sat, 28 Sep 2013 21:06:12 -0400 Morris Meyer PTX single-threaded array store, Warp annotation
Sat, 14 Sep 2013 17:31:25 -0400 Morris Meyer Start of PTX array passing
Wed, 11 Sep 2013 17:02:40 -0400 bharadwaj Create runtime object based on GPUIL type newly added to gpu class; some code refactoring.
Tue, 10 Sep 2013 22:39:50 -0400 bharadwaj Initial implementation of PTXRuntime (RegisterConfig, PTX description etc); guarded with new flag UseGPU. Specify -XX:+UseGPU to exercise this new implementation.
Wed, 04 Sep 2013 10:47:37 -0400 bharadwaj Improvements to PTX codegen; allows more PTX tests that run on the device to pass.
Fri, 30 Aug 2013 16:39:05 -0400 bharadwaj Implement several missing PTX codegen features; return value capture and method args passing of java method executed on GPU.
Sat, 10 Aug 2013 22:57:50 +0200 Bernhard Urban ptx: fix include for non-precompiledheader build in gpu_ptx.cpp
Sat, 10 Aug 2013 14:38:30 -0400 Morris Meyer CR-20: PTX kernel invocation with arguments - from Bharadwaj
Fri, 26 Jul 2013 12:09:17 +0200 Roland Schatz Remove CUDA message at vm startup.
Fri, 26 Jul 2013 11:47:38 +0200 Roland Schatz Fix compilation without precompiled headers.
Fri, 26 Jul 2013 11:22:13 +0200 Christian Wirth fix: gpu_ptx.cpp did not compile on windows
Thu, 25 Jul 2013 22:15:30 -0400 Morris Meyer PTX support for Linux