summaryrefslogtreecommitdiff
path: root/src/cl_device_id.h
AgeCommit message (Collapse)AuthorFilesLines
2015-10-21Runtime: add CL_DEVICE_SPIR_VERSIONS to clGetDeviceInfo.Yang Rong1-0/+2
Signed-off-by: Yang Rong <rong.r.yang@intel.com> Reviewed-by: Luo Xionghu <xionghu.luo@intel.com>
2015-09-22enable create image 2d from buffer in clCreateImage.Luo Xionghu1-0/+2
this patch allows create 2d image with a cl buffer with zero copy. v2: should use reference to manage the release the buffer and image. After being created, the buffer reference count is 2, and image reference count is 1. if image is released first, decrease the image reference count and buffer reference count both, release the bo when the buffer is released at last; if buffer is released first, decrease the buffer reference count only, release the buffer when the image is released. add CL_DEVICE_IMAGE_BASE_ADDRESS_ALIGNMENT in cl_device_info. v3: move is_image_from_buffer to _cl_mem_image; return CL_INVALID_IMAGE_SIZE if image size is larger than the buffer. v4: pitchalignment set to 2. Signed-off-by: Luo Xionghu <xionghu.luo@intel.com> Reviewed-by: Guo, Yejun <yejun.guo@intel.com>
2015-08-13Set proper Vendor IDMidhun Kodiyath1-0/+1
Device ID and vendor ID are not same.Set the correct vendor ID. Signed-off-by: Midhun Kodiyath <midhunchandra.kodiyath@intel.com> Reviewed-by: Song, Ruiling <ruiling.song@intel.com> Reviewed-by: Guo, Yejun <yejun.guo@intel.com>
2015-07-06runtime: Add cl device's standalone extension.Junyan He1-1/+1
The cl device may have different extensions from the platform. We will add some items based on the platform extensions. Signed-off-by: Junyan He <junyan.he@linux.intel.com> Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>
2015-07-02runtime: Add fp16 extension to BDW later platform.Junyan He1-0/+1
Signed-off-by: Junyan He <junyan.he@linux.intel.com> Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>
2015-06-30use self test to determine enable/or disable atomics in L3 for HSW.Luo Xionghu1-0/+1
check the selftest kernel return value, if enqueue kernel failed, set the flag to not enable atomics the L3 for HSW. This reverts commit 83f8739b6fc4893fac60145326052ccb5cf653dc. v2: don't use global variable to pass value from runtime to driver. v3: add type SELF_TEST_OTHER_FAIL to differentiate from SELF_TEST_ATOMIC_FAIL; seperate the ATOMIC_FAIL from SLM_FAIL, only SLM_FAIL can be control by env OCL_IGNORE_SELF_TEST. Signed-off-by: Luo Xionghu <xionghu.luo@intel.com> Reviewed-by: Yang, Rong <rong.r.yang@intel.com>
2014-11-11License: adjust all license version to LGPL v2.1+.Zhigang Gong1-1/+1
To make the license statement consistent to each other, adjust all license versions to v2.1+. Thus beignet should have a pure LGPL v2.1+ license. Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
2014-10-27GBE: fix a wrong type of cl_device_info.Zhigang Gong1-1/+1
Per OpenCL spec 1.2: CL_DEVICE_IMAGE_MAX_BUFFER_SIZE should be size_t type rather than cl_ulong. This bug will cause problems on i386 platform. Signed-off-by: Zhigang Gong <zhigang.gong@intel.com> Tested-by: "Meng, Mengmeng" <mengmeng.meng@intel.com>
2014-10-10BDW: Add device's sub slice field, for cl_get_kernel_max_wg_sz.Yang Rong1-0/+1
When SLM enable, get kernal max workgroup size should return the a sub slice's max thread * simdwidth. So need the sub slice information. Signed-off-by: Yang Rong <rong.r.yang@intel.com> Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com> Reviewed-by: Junyan He <junyan.he@linux.intel.com>
2014-09-11fix clGetKernelWorkGroupInfo built-in kernel fail.Luo Xionghu1-0/+3
add CL_KERNEL_GLOBAL_WORK_SIZE option for clGetKernelWorkGroupInfo. v2: should return the max global work size instead of current work size. This funtion need return CL_INVALID_VALUE if the device is not a custom device or kernel is not a built-in kernel. we have 3 kind of built-in kernels for 1d/2d/3d memories, the max global work size are decided by the dimension and memory type. the piglit fail is caused by calling NON built-in kernels, so need send patch to piglit later. Signed-off-by: Luo Xionghu <xionghu.luo@intel.com> Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2014-07-01runtime: fix max group size calculation issue.Zhigang Gong1-6/+6
If the kernel doesn't use slm/barrier, there is no hard limitation for the max group size. And if the max work group size is more than 1024, the original 64 urb entry count will not be sufficient to hold all the curbe payload. Change the entry count to max thread count to fix this potential issue. I found this bug when I tried to run phoronix test suite's juliagpu test case on my MBA. v2: refine the max kernel work group size calculation mechanism. the wg_sz should not be a device's member variable, it should be a variable derived from kernel and device's attriute at runtime. also fix wrong configuration for IVB GT1. v3: Add an important max thread limitation in the GPGPU_WALKER command. For non-Baytrail, the max thread depth * max thread height * max thread width should less than 64 (under either simd16 or simd8), no matter whether SLM/barrier is used. We oversighted that limitation before, thus for a simd8 kernel which use work group size 1024 will exceed this limitation and half of the thread will not be executed at all. Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com> Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>
2014-06-23Add some OpenCL1.2 parameters of function clGetDeviceInfo.Yang Rong1-0/+3
Include CL_DEVICE_LINKER_AVAILABLE, CL_DEVICE_PRINTF_BUFFER_SIZE, CL_DEVICE_PREFERRED_INTEROP_USER_SYNC. Signed-off-by: Yang Rong <rong.r.yang@intel.com> Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2014-06-20Add some API's OpenCL 1.2 parameter support.Yang Rong1-0/+1
Support CL_PROGRAM_KERNEL_NAMES and CL_PROGRAM_NUM_KERNELS in API clGetProgramInfo, and CL_DOUBLE_FP_CONFIG in API clGetDeviceInfo. Also fix a bug of CL_MEM_HOST_PTR in API clGetMemObjectInfo. v2: also fix the utest get_mem_info. Signed-off-by: Yang Rong <rong.r.yang@intel.com> Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2014-06-20GBE/runtime: fixup broken 1d array image support.Zhigang Gong1-0/+1
As sample LD message doesn't support array index, we have to create a 2D array surface with the same buffer object. Thus one 1D array image will have two surfaces binded to it one is the index and the second is 128 + index. And then at kernel side, we will access the corresponding 2D array surface when the LD message is required otherwise will access the origin 1D array surface. Signed-off-by: Zhigang Gong <zhigang.gong@intel.com> Reviewed-by: He Junyan <junyan.he@inbox.com>
2014-06-13add [opencl-1.2] API clCreateSubDevice.Luo1-0/+7
creates an array of sub-devices that each reference a non-intersecting set of compute units within in_device, according to a partition scheme given by properties. Reviewed-by: He Junyan <junyan.he@inbox.com> Signed-off-by: Luo <xionghu.luo@intel.com>
2014-06-13Add the support for 1D image from buffer.Junyan He1-0/+1
Signed-off-by: Junyan He <junyan.he@linux.intel.com> Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2014-02-17CL: make the scratch size as a device resource attribute.Zhigang Gong1-0/+1
Actually, the scratch size is much like the local memory size which should be a device dependent information. This patch is to put scratch mem size to the device attribute structure. And when the kernel needs more than the maximum scratch memory, we just return a out-of-resource error rather than trigger an assertion. Signed-off-by: Zhigang Gong <zhigang.gong@intel.com> Reviewed-by: Song, Ruiling <ruiling.song@intel.com>
2013-11-13GBE: Add support for kernel attribute reqd_work_group_size.Zhigang Gong1-1/+0
When a kernel has __attribute__((reqd_work_group_size(X, Y, Z))) qualifier, the kernel will only accept that group size. v2: add binary load/store support. v3: fix the MDNode parsing according to spir spec. It's using the following structure rather than a tbaa tree. !spir.functions = !f !0,!1,...,!N g ; Note: The first element is always an LLVM::Function signature !0 = metadata !f < function signature >, !01, !02, ..., , !0i g !1 = metadata !f < function signature >, !11, !12, ..., , !1j g ... !N = metadata !f < function signature >, !N1, !N2, ..., , !Nk g Signed-off-by: Zhigang Gong <zhigang.gong@intel.com> Reviewed-by: "Yang, Rong R" <rong.r.yang@intel.com>
2013-11-11Runtime: complete the api clGetKernelWorkGroupInfo.Zhigang Gong1-5/+7
Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
2013-05-17Correct type of device propertiesSimon Richter1-2/+2
- CL_DEVICE_MAX_PARAMETER_SIZE is of type size_t - CL_DEVICE_MAX_WORK_GROUP_SIZE is of type size_t Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2013-05-17Add clGetDeviceInfo(..., CL_BUILT_IN_KERNELS, ...)Simon Richter1-0/+2
Currently, there are no built-in kernels, so this function returns an empty string. Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2013-04-24Implement clGetDeviceInfo(..., CL_DRIVER_VERSION, ...)Simon Richter1-0/+2
This returns the library major/minor version. As it does not follow the usual naming scheme, the output code is duplicated. Signed-off-by: Simon Richter <Simon.Richter@hogyros.de> Reviewed-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2013-04-18Implement KHR ICD extensionSimon Richter1-0/+1
This adds a pointer to the dispatch table at the beginning of every object of type - cl_command_queue - cl_context - cl_device_id - cl_event - cl_kernel - cl_mem - cl_platform_id - cl_program - cl_sampler as required by the ICD specification. The layout of the dispatch table comes from the OpenCL ICD loader by Brice Videau <brice.videau@imag.fr> and Vincent Danjean <Vincent.Danjean@ens-lyon.org>. To avoid dispatch table entries being overwritten with the ICD loader's implementations of the CL functions (as would be the proper behaviour for the ELF loader), the -Bsymbolic option is given to the linker. Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2012-08-10Cleaned up the code from previous unused dependencies Upate the READMEBenjamin Segovia1-2/+1
2012-08-10Added first test for stack handlingBenjamin Segovia1-0/+1
2012-08-10Added support for: clEnqueueReadBuffer clgetkernelWorkGroupInfo: For this ↵hlthantr1-0/+8
currently, a static solution based on the type of device has been implemented. This has been done for the sake of completeness. A real implementation should ideally parse the kernel, and extract this information. Fixed issue: A return value issue in drm_intel_bo_subdata, where different versions of the library differ in how they treat error. In one case, an rval of zero indicates success, and in the other, it indicates failure. Fix is to remove the checking of rval entirely.
2012-08-10Cleaned and simplified all test kernels and codes. Now everything is inside ↵bsegovia1-0/+3
one reportory and kernels are compiled and sorted per generation
2012-08-10Added all miniCL filesbsegovia1-0/+111
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-12 05:38:41 +0000