Age | Commit message (Collapse) | Author | Files | Lines |
|
create a w* (3/2*h) size bo for the whole CL_NV12_INTEL format
surface, and the y surface (format CL_R) share the first w * h
part, uv surface (format CL_RG) share the left w * 1/2h part; set
correct bo offset for uv surface per different platforms.
v2: add extension define in libocl; fix error check.
Signed-off-by: Luo Xionghu <xionghu.luo@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
SVM enqueues need to call cl_event_exec every time.
Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Junyan He <junyan.he@linux.intel.com>
|
|
cl_event_exec it the uniformal entry for all event command execution,
call cl_enqueue_handle may miss time stamp record.
Replace all cl_enqueue_handle to cl_event_exec.
Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Pan Xiuli <xiuli.pan@intel.com>
|
|
Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: Ruiling Song <ruiling.song@intel.com>
|
|
TODO:
In opencl 2.0, a new profiling item called CL_PROFILING_COMMAND_COMPLETE
is imported. It means that we need to record the time stamp of all the
child events created by the "Kernel enqueing kernels" feature finish.
This should be done after the "Kernel enqueing kernels" feature enabled.
V2:
Update event time stamp before inserting to queue thread, avoid MT issue.
V3:
Fixup overflow problem.
V4:
Fixup overflow to 0xfffffffffffffffff problem.
Just take ownership and release event lock when call the update timestamp
function. The update timestamp function may have block system call can
should not hold the lock to call it.
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Also add some parameters check to this API.
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
In future there may be more than one device on the platform,
we need to consider multi devices within one context.
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Add multi-device logic, make it more readable
and move it to new file.
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
V2:
Delete some useless macro.
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
V2: Remove check for device queue and add device queue flag.
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Reviewed-by: Pan Xiuli <xiuli.pan@intel.com>
|
|
sRGB writes are not supported now, and we should not return them
if any write was set as CL_MEM_FLAGS.
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
CL_sRGBA with CL_UNNORM_INT8 is the minimum request for OpenCL2.0
and CL_sBGRA is also support by hardware.
None of the sRGB surface type suport hardware write.
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Add clCreatePipe and clGetPipeInfo
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Add api clCreateSamplerWithProperties
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
The extra exec info need reloc, otherwize gpu can't read/write.
And it don't need set to curbe.
So reloc it to unused binding table.
Signed-off-by: Yang Rong <rong.r.yang at intel.com>
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Pan Xiuli <xiuli.pan@intel.com>
|
|
Contributor: Luo Xionghu <xionghu.luo at intel.com>
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>
|
|
Enable CL_DEVICE_SVM_COARSE_GRAIN_BUFFER svm support, use userptr and softpin
to implement it. Use userptr to share the page between cpu and gpu, and softpin
to unify the cpu and gpu's address. Now it works on i386 system. x86_64 depends on
backend support.
This patch base on DRM library and DRM kernel driver's softpin patch:
http://lists.freedesktop.org/archives/intel-gfx/2015-September/075446.html.
Signed-off-by: Yang Rong <rong.r.yang@intel.com>
Reviewed-by: "Song, Ruiling" <ruiling.song@intel.com>
|
|
V2:
Move the event list status check to clWaitForEvents API.
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Rewrite the cl_event, and modify all the event functions
using this new event manner. Event will co-operate with
command queue's thread together.
v2:
Fix a logic problem in event create failed.
V3:
Set enqueue default to do nothing, handle some enqueue has nothing
to do.
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Junyan He <junyan.he@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
local_size
Signed-off-by: Guo Yejun <yejun.guo@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Add clGetKernelSubGroupInfoKHR in the clGetExtensionFunctionAddress
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Pan Xiuli <xiuli.pan@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
this patch fixed the issue at https://bugs.freedesktop.org/show_bug.cgi?id=95136
Signed-off-by: Guo Yejun <yejun.guo@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
If the user provides local_work_size as NULL in clEnqueueNDRangeKernel,
and we could not find a good value inside driver, output a warning
message with macro DEBUGP, and also refine the macro.
Signed-off-by: Guo Yejun <yejun.guo@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
this link fail appears on gcc 5.2.1.
Signed-off-by: Luo Xionghu <xionghu.luo@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|
|
Signed-off-by: Luo Xionghu <xionghu.luo@intel.com>
Reviewed-by: Yang Rong <rong.r.yang@intel.com>
|