From: Luo Xionghu xionghu@intel.com
We can change the image_channel_order to CL_RGBA and
image_channel_data_type to CL_UNSIGNED_INT32 for some special
case, thus 16 bytes can be read by one work item. Bandwidth is
fully used.
v2: merge patch 3 of initializing region0;
remove k dimension in
On 2015.04.08 14:50:47 +0800, Chuanbo Weng wrote:
This example captures yuy2 frame directly to cl buffer object by the way
of dma, processed by OpenCL kernel, then convert to nv12 format and
shown by libva.
One thing confusing here is that for current beignet's clGetMemObjectFdIntel(),
you
Some of these register is request as scalar ir::register and retype to uniform
GenRegister.
These temp registers are all used as send's address or other payload, the send
payload need continuous register, so I change these registers to scalar
register.
-Original Message-
From:
OK, good idea, I will change the default configure and add the warning.
-Original Message-
From: Mcgee, Jeff
Sent: Saturday, April 4, 2015 03:46
To: Zhigang Gong
Cc: Yang, Rong R; beignet@lists.freedesktop.org
Subject: Re: [Beignet] [PATCH 1/2] CHV: Add cherryview support in the
From: Luo Xionghu xionghu@intel.com
v2: use random input data; update comments.
v3: change the image attribute to __write_only.
Signed-off-by: Luo Xionghu xionghu@intel.com
---
benchmark/CMakeLists.txt | 1 +
benchmark/benchmark_copy_buffer_to_image.cpp |
On Tue, Mar 31, 2015 at 04:39:03PM +0800, Yang Rong wrote:
There is a segmentation fault in function isSrcDstDiffSpan, when src's hstrde
is not GEN_HORIZONTAL_STRIDE_0 but dst's hstride is GEN_HORIZONTAL_STRIDE_0.
This is wrong state, and the LoadInstruction using GenRegister::udxgrf with
This example captures yuy2 frame directly to cl buffer object by the way
of dma, processed by OpenCL kernel, then convert to nv12 format and
shown by libva.
v2:
Close cl buffer's fd by clCloseMemObjectFdIntel instead of close
function.
v3:
Just use close function, no need of
with a 4.0.0-rc6 kernel
...which includes that fix as standard.
Has anyone else tried beignet on Haswell-ULT, and if so, did it work?
___
Beignet mailing list
Beignet@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/beignet
In general I agree with you. We will clean up and refine these extensions later.
I've sent out a new version of this patchset and ask Yuan Feng for review.
Thanks,
Chuanbo Weng
-Original Message-
From: Zhigang Gong [mailto:zhigang.g...@linux.intel.com]
Sent: Friday, April 03, 2015 9:45
ok. As we discussed, you'd better send out a new version:)
-Original Message-
From: Luo, Xionghu
Sent: Friday, April 03, 2015 11:17
To: Weng, Chuanbo; beignet@lists.freedesktop.org
Subject: RE: [Beignet] [patch v2 1/2] Optimization of
clEnqueueCopyBufferToImage for 16 aligned case.
I
Ah yes, echo 0 /sys/module/i915/parameters/enable_cmd_parser fixed
everything. Now all the tests pass and other OpenCL using programs work
great! If I get a chance I will spend the time to try and bisect which
kernel version causes issues.
On 04/08/2015 11:04 AM, Zhigang Gong wrote:
We tested
11 matches
Mail list logo
