Hi Andreas,
The driver information is as follows:
filename:
/lib/modules/3.13.0-61-generic/kernel/drivers/video/nvidia.ko
alias: char-major-195-*
version: 352.30
supported: external
license: NVIDIA
alias: pci:v000010DEd00000E00sv*sd*bc04sc80i00*
alias: pci:v000010DEd*sv*sd*bc03sc02i00*
alias: pci:v000010DEd*sv*sd*bc03sc00i00*
depends: drm
vermagic: 3.13.0-61-generic SMP mod_unload modversions
parm: NVreg_Mobile:int
parm: NVreg_ResmanDebugLevel:int
parm: NVreg_RmLogonRC:int
parm: NVreg_ModifyDeviceFiles:int
parm: NVreg_DeviceFileUID:int
parm: NVreg_DeviceFileGID:int
parm: NVreg_DeviceFileMode:int
parm: NVreg_UpdateMemoryTypes:int
parm: NVreg_InitializeSystemMemoryAllocations:int
parm: NVreg_UsePageAttributeTable:int
parm: NVreg_MapRegistersEarly:int
parm: NVreg_RegisterForACPIEvents:int
parm: NVreg_CheckPCIConfigSpace:int
parm: NVreg_EnablePCIeGen3:int
parm: NVreg_EnableMSI:int
parm: NVreg_MemoryPoolSize:int
parm: NVreg_RegistryDwords:charp
parm: NVreg_RmMsg:charp
parm: NVreg_AssignGpus:charp
Using only one queue is not an option, as more than one device has to be
used with advanced syncing
and therefore the event system should be in a usable state.
Thanks for any help.
Jonathan
Am 2016-04-05 21:59, schrieb Andreas Kloeckner:
Hi Jonathan,
"Schock, Jonathan" <jonathan.sch...@tum.de> writes:
The important bits that seem to the behaviour are:
import pyopencl as cl
import numpy as np
platform = cl.get_platforms()[0]
devs = platform.get_devices()
device1 = devs[0]
h_data =
np.arange(81*9000).reshape(90,90,90).astype(np.float32,order='F')
ctx = cl.Context([device1])
queue = cl.CommandQueue(ctx)
queue2 = cl.CommandQueue(ctx)
h_image_shape=h_data.shape
f = open('Kernel.cl', 'r')
fstr = "".join(f.readlines())
prg = cl.Program(ctx, fstr).build()
d_image = cl.Image(ctx, mf.READ_ONLY,
cl.ImageFormat(cl.channel_order.INTENSITY,cl.channel_type.FLOAT),h_image_shape)
wev1 = cl.enqueue_copy(queue, d_image, h_data, is_blocking=False,
origin=(0,0,0), region=h_image_shape)
prg.sum(queue2,(h_image_shape[0],),None,d_image,wait_for = [wev1])
The Kernel is doing some simple number crunching on the input image.
I'm measuring with nvvp, one result is attached, where you clearly
see,
that the kernel launches long before the copy has ended.
I allready tested with OoO disabled...same behaviour.
Implementation is 'Tesla K10.G2.8GB' on 'NVIDIA CUDA'.
For completeness: What driver version is this?
I am pretty certain that what you're seeing is non-compliant behavior
by
the Nv implementation. The behavior you're seeing is consistent with Nv
just ignoring the event_wait_list.
Beyond that, using two queues *and* ooo is redundant. ooo alone will
already allow concurrency.
https://www.khronos.org/registry/cl/sdk/1.2/docs/man/xhtml/clCreateCommandQueue.html
I would probably recommend using a single queue until overlapping
becomes crucial.
Andreas
_______________________________________________
PyOpenCL mailing list
PyOpenCL@tiker.net
https://lists.tiker.net/listinfo/pyopencl
