[Beignet] troubles with beignet 0.8.0, Debian 8 (jessie) and i7 4770

Yang, Rong R rong.r.yang at intel.com
Thu Jun 12 19:41:10 PDT 2014



-----Original Message-----
From: Beignet [mailto:beignet-bounces at lists.freedesktop.org] On Behalf Of Yichao Yu
Sent: Friday, June 13, 2014 2:09 AM
To: Nikita Orlov
Cc: beignet at lists.freedesktop.org
Subject: Re: [Beignet] troubles with beignet 0.8.0, Debian 8 (jessie) and i7 4770

On Thu, Jun 12, 2014 at 12:52 PM, Nikita Orlov <nikitos1550 at yandex.ru> wrote:
> Hello!
>
> I`m trying to run opencv opencl accelerated.
> I have installed beignet from debian jessie repo.
>
> clinfo:
>
> user at video-stend2:~$ clinfo
> Number of platforms: 1
>   Platform Profile: FULL_PROFILE
>   Platform Version: OpenCL 1.1 beignet 0.8.0
>   Platform Name: Experiment Intel Gen OCL Driver
>   Platform Vendor: Intel
>   Platform Extensions: cl_khr_global_int32_base_atomics 
> cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics 
> cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store 
> cl_khr_icd
>
>
>   Platform Name: Experiment Intel Gen OCL Driver Number of devices: 1
>   Device Type: CL_DEVICE_TYPE_GPU
>   Device ID: 1042
>   Max compute units: 64
>   Max work items dimensions: 3
>     Max work items[0]: 512
>     Max work items[1]: 512
>     Max work items[2]: 512
>   Max work group size: 512
>   Preferred vector width char: 16
>   Preferred vector width short: 16
>   Preferred vector width int: 16
>   Preferred vector width long: 16
>   Preferred vector width float: 16
>   Preferred vector width double: 0
>   Native vector width char: 16
>   Native vector width short: 16
>   Native vector width int: 16
>   Native vector width long: 16
>   Native vector width float: 16
>   Native vector width double: 16
>   Max clock frequency: 1000Mhz
>   Address bits: 32
>   Max memory allocation: 134217728
>   Image support: Yes
>   Max number of images read arguments: 128
>   Max number of images write arguments: 8
>   Max image 2D width: 8192
>   Max image 2D height: 8192
>   Max image 3D width: 8192
>   Max image 3D height: 8192
>   Max image 3D depth: 2048
>   Max samplers within kernel: 16
>   Max size of kernel argument: 1024
>   Alignment (bits) of base address: 1024
>   Minimum alignment (bytes) for any datatype: 128
>   Single precision floating point capability
>     Denorms: No
>     Quiet NaNs: Yes
>     Round to nearest even: Yes
>     Round to zero: No
>     Round to +ve and infinity: No
>     IEEE754-2008 fused multiply-add: No
>   Cache type: Read/Write
>   Cache line size: 128
>   Cache size: 8192
>   Global memory size: 134217728
>   Constant buffer size: 524288
>   Max number of constant args: 8
>   Local memory type: Global
>   Local memory size: 65536
>   Error correction support: 0
>   Unified memory for Host and Device: 0
>   Profiling timer resolution: 80
>   Device endianess: Little
>   Available: Yes
>   Compiler available: Yes
>   Execution capabilities:
>     Execute OpenCL kernels: Yes
>     Execute native function: Yes
>   Queue properties:
>     Out-of-Order: No
>     Profiling : Yes
>   Platform ID: 0x7f1390957a40
>   Name: Intel(R) HD Graphics Haswell M
>   Vendor: Intel
>   Device OpenCL C version: OpenCL C 1.1 beignet 0.8.0
>   Driver version: 0.8.0
>   Profile: FULL_PROFILE
>   Version: OpenCL 1.1 beignet 0.8.0
>   Extensions: cl_khr_global_int32_base_atomics 
> cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics 
> cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store 
> cl_khr_icd
>
>
> user at video-stend2:~$ clinfo
> Number of platforms: 1
>   Platform Profile: FULL_PROFILE
>   Platform Version: OpenCL 1.1 beignet 0.8.0
>   Platform Name: Experiment Intel Gen OCL Driver
>   Platform Vendor: Intel
>   Platform Extensions: cl_khr_global_int32_base_atomics 
> cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics 
> cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store 
> cl_khr_icd
>
>
>   Platform Name: Experiment Intel Gen OCL Driver Number of devices: 1
>   Device Type: CL_DEVICE_TYPE_GPU
>   Device ID: 1042
>   Max compute units: 64
>   Max work items dimensions: 3
>     Max work items[0]: 512
>     Max work items[1]: 512
>     Max work items[2]: 512
>   Max work group size: 512
>   Preferred vector width char: 16
>   Preferred vector width short: 16
>   Preferred vector width int: 16
>   Preferred vector width long: 16
>   Preferred vector width float: 16
>   Preferred vector width double: 0
>   Native vector width char: 16
>   Native vector width short: 16
>   Native vector width int: 16
>   Native vector width long: 16
>   Native vector width float: 16
>   Native vector width double: 16
>   Max clock frequency: 1000Mhz
>   Address bits: 32
>   Max memory allocation: 134217728
>   Image support: Yes
>   Max number of images read arguments: 128
>   Max number of images write arguments: 8
>   Max image 2D width: 8192
>   Max image 2D height: 8192
>   Max image 3D width: 8192
>   Max image 3D height: 8192
>   Max image 3D depth: 2048
>   Max samplers within kernel: 16
>   Max size of kernel argument: 1024
>   Alignment (bits) of base address: 1024
>   Minimum alignment (bytes) for any datatype: 128
>   Single precision floating point capability
>     Denorms: No
>     Quiet NaNs: Yes
>     Round to nearest even: Yes
>     Round to zero: No
>     Round to +ve and infinity: No
>     IEEE754-2008 fused multiply-add: No
>   Cache type: Read/Write
>   Cache line size: 128
>   Cache size: 8192
>   Global memory size: 134217728
>   Constant buffer size: 524288
>   Max number of constant args: 8
>   Local memory type: Global
>   Local memory size: 65536
>   Error correction support: 0
>   Unified memory for Host and Device: 0
>   Profiling timer resolution: 80
>   Device endianess: Little
>   Available: Yes
>   Compiler available: Yes
>   Execution capabilities:
>     Execute OpenCL kernels: Yes
>     Execute native function: Yes
>   Queue properties:
>     Out-of-Order: No
>     Profiling : Yes
>   Platform ID: 0x7f1390957a40
>   Name: Intel(R) HD Graphics Haswell M
>   Vendor: Intel
>   Device OpenCL C version: OpenCL C 1.1 beignet 0.8.0
>   Driver version: 0.8.0
>   Profile: FULL_PROFILE
>   Version: OpenCL 1.1 beignet 0.8.0
>   Extensions: cl_khr_global_int32_base_atomics 
> cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics 
> cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store 
> cl_khr_icd ccc
>
> I am trying run opencv_test_ocl and utest (from beignet src, I can succesfully build ut by myself). Everything fails.
> It is difficult to paste here log, because system freeze and I can just reset computer.

It seems that you have a Haswell CPU and I think you are having the same problem with me (You can find a thread on the mailing list a few weeks ago about it). The support for Haswell CPU is not finished yet and it will very likely hang the GPU.

>>>>> Are you using beignet 0.8 release? Beignet v0.8 have not supported HSW, you should use Beignet git master, the Haswell support almost has finished in master. Pay attention that you should apply a kernel patch in attachment if you want to run kernels with local memory. This is because of drm's restriction, and will drm guys is handing it. 
If still any problem, welcome to report.

P.S. I don't think the system is frozen. It might appear that way because the display is not updating. However, if you wait long enough and/or try to kill the running process with Ctrl-C, you can still get you system back.

>>>>>>System frozen may be caused by GPU hang and your system doesn't enable hang check. Try echo -n 1 > /sys/module/i915/parameters/enable_hangcheck with root user.

>
> I run tests with Xorg and without.
>
> What is wrong? Maybe I didn`t install some libs or some additional configuration is required?
>
> --
> Nikita Orlov
> Skype: nik_stet
> QQ: 2717846083
> _______________________________________________
> Beignet mailing list
> Beignet at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/beignet
_______________________________________________
Beignet mailing list
Beignet at lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/beignet
-------------- next part --------------
A non-text attachment was scrubbed...
Name: disable-batchbuffer-security.patch
Type: application/octet-stream
Size: 497 bytes
Desc: disable-batchbuffer-security.patch
URL: <http://lists.freedesktop.org/archives/beignet/attachments/20140613/dcd1e1e8/attachment-0001.obj>


More information about the Beignet mailing list