找回密码
 新注册用户
搜索
查看: 1965|回复: 8

[求助] PrimeGrid GPU任务报错

[复制链接]
发表于 2022-3-4 11:17:33 | 显示全部楼层 |阅读模式
本帖最后由 samchoi 于 2022-3-4 11:34 编辑

抱歉了各位,每次来论坛都是求助

GPU是GTX260,驱动版本342.01,Win10,6G内存

具体报错信息如下:一是Wieferich and Wall-Sun-Sun的
用户名        ww_3840993_0
任务单元        780038649
创建时间        3 Mar 2022 | 15:28:22 UTC
发送时间        3 Mar 2022 | 17:44:38 UTC
接收时间        3 Mar 2022 | 19:04:59 UTC
服务端状态        完成
结果        计算错误
客户端状态        计算错误
退出状态        1 (0x00000001) Unknown error code
计算机 ID        1124568
上报期限        11 Mar 2022 | 18:44:38 UTC
运行时间        0.00
CPU 时间        0.00
验证状态        无效结果
积分        0.00
计算程序版本        Wieferich and Wall-Sun-Sun v1.05 (opencl_nvidia_WW)
标准错误输出
<core_client_version>7.16.20</core_client_version>
<![CDATA[
<message>
- exit code 1 (0x1)</message>
<stderr_txt>

WWocl version 1.04 by Bryan Little and Yves Gallot
Compiled Dec 31 2020 with GCC 8.1.0
A Wieferich and WallSunSun prime number search program
Based on wwwwcl by Mark Rodenkirch and primesieve by Kim Walisch
Command line: projects/www.primegrid.com/WWocl-1.04-win64.exe 10712218000000000000 10712221000000000000
Starting search at: 10712218000000000000
Stopping search at: 10712221000000000000
GPU Info:
  Name:                 GeForce GTX 260
  Vendor:                 NVIDIA Corporation
  Driver:                 342.01
  Compute Units:         24
ERROR: clBuildProgram failure for wieferich
Build Log for wieferich_program:
ptxas application ptx input, line 201; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 204; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas application ptx input, line 207; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 210; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas application ptx input, line 213; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 216; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas application ptx input, line 294; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 297; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas application ptx input, line 300; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 303; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas application ptx input, line 306; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 309; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas application ptx input, line 334; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 337; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas application ptx input, line 340; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 343; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas application ptx input, line 346; error   : Instruction 'mad.cc' requires .target sm_20 or higher
ptxas application ptx input, line 349; error   : Instruction 'madc' requires .target sm_20 or higher
ptxas fatal   : Ptx assembly aborted due to errors


CL_INVALID_BINARY

</stderr_txt>
]]>
回复

使用道具 举报

 楼主| 发表于 2022-3-4 11:20:30 | 显示全部楼层
二是Genefer 18 v3.21 (OCLcudaGFN18)(中间有一大段好像是重复的就没帖上来)


用户名        genefer18_60718260_1
任务单元        779987594
创建时间        3 Mar 2022 | 10:03:49 UTC
发送时间        3 Mar 2022 | 19:04:59 UTC
接收时间        3 Mar 2022 | 20:07:19 UTC
服务端状态        完成
结果        计算错误
客户端状态        计算错误
退出状态        20 (0x00000014) Unknown error code
计算机 ID        1124568
上报期限        8 Mar 2022 | 20:04:59 UTC
运行时间        2.24
CPU 时间        0.16
验证状态        无效结果
积分        0.00
计算程序版本        Genefer 18 v3.21 (OCLcudaGFN18)
标准错误输出
<core_client_version>7.16.20</core_client_version>
<![CDATA[
<message>
- exit code 20 (0x14)</message>
<stderr_txt>geneferocl 3.3.3-2 (Windows/OpenCL/32-bit)

Copyright 2001-2018, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2014, Michael Goetz, Ronald Schneider
Copyright 2011-2018, Iain Bethune
Genefer is free source code, under the MIT license.

Running on platform 'NVIDIA CUDA', device 'GeForce GTX 260', vendor 'NVIDIA Corporation', version 'OpenCL 1.0 CUDA' and driver '342.01'.
24 computeUnits @ 1242MHz, memSize=896MB, cacheSize=0kB, cacheLineSize=0B, localMemSize=16kB, maxWorkGroupSize=512.
Supported transform implementations: ocl ocl2 ocl3 ocl4 ocl5

Command line: projects/www.primegrid.com/geneferocl_windows_3.3.3-2.exe -boinc -q 15135866^262144+1

Normal priority change succeeded.
Using OCL2 transform
Error: OpenCL error detected: CL_INVALID_BINARY.
Errors occurred for all available transform implementations
Waiting 10 minutes before attempting to continue from last checkpoint...

geneferocl 3.3.3-2 (Windows/OpenCL/32-bit)

Copyright 2001-2018, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2014, Michael Goetz, Ronald Schneider
Copyright 2011-2018, Iain Bethune
Genefer is free source code, under the MIT license.

Running on platform 'NVIDIA CUDA', device 'GeForce GTX 260', vendor 'NVIDIA Corporation', version 'OpenCL 1.0 CUDA' and driver '342.01'.
24 computeUnits @ 1242MHz, memSize=896MB, cacheSize=0kB, cacheLineSize=0B, localMemSize=16kB, maxWorkGroupSize=512.
Supported transform implementations: ocl ocl2 ocl3 ocl4 ocl5

Command line: projects/www.primegrid.com/geneferocl_windows_3.3.3-2.exe -boinc -q 15135866^262144+1

Normal priority change succeeded.
Calculation failed after 6 tries, please test your GPU or CPU carefully to ensure it can reliably compute GFN tests.
04:05:39 (10160): called boinc_finish(20)

</stderr_txt>
]]>
回复

使用道具 举报

 楼主| 发表于 2022-3-4 11:22:05 | 显示全部楼层
第三是Genefer 21 v3.21 (OCLcudaGFN)

<core_client_version>7.16.20</core_client_version>
<![CDATA[
<message>
- exit code 20 (0x14)</message>
<stderr_txt>
geneferocl 3.3.3-2 (Windows/OpenCL/32-bit)

Copyright 2001-2018, Yves Gallot
Copyright 2009, Mark Rodenkirch, David Underbakke
Copyright 2010-2012, Shoichiro Yamada, Ken Brazier
Copyright 2011-2014, Michael Goetz, Ronald Schneider
Copyright 2011-2018, Iain Bethune
Genefer is free source code, under the MIT license.

Running on platform 'NVIDIA CUDA', device 'GeForce GTX 260', vendor 'NVIDIA Corporation', version 'OpenCL 1.0 CUDA' and driver '341.74'.
24 computeUnits @ 1242MHz, memSize=896MB, cacheSize=0kB, cacheLineSize=0B, localMemSize=16kB, maxWorkGroupSize=512.
Supported transform implementations: ocl ocl2 ocl3 ocl4 ocl5

Command line: projects/www.primegrid.com/geneferocl_windows_3.3.3-2.exe -boinc -q 788372^2097152+1

Normal priority change succeeded.
Checking available transform implementations...
A benchmark is needed to determine best transform, testing available transform implementations...
Testing OCL transform...
Error: OpenCL error detected: CL_INVALID_BINARY.
Testing OCL2 transform...
Error: OpenCL error detected: CL_INVALID_BINARY.
Testing OCL3 transform...
Error: OpenCL error detected: CL_INVALID_BINARY.
Testing OCL4 transform...
Error: OpenCL error detected: CL_INVALID_BINARY.
Benchmarks completed (20.497 seconds).
Using OCL5 transform
Error: OpenCL error detected: CL_INVALID_BINARY.
Errors occurred for all available transform implementations
Waiting 10 minutes before attempting to continue from last checkpoint...

回复

使用道具 举报

发表于 2022-3-10 14:16:52 | 显示全部楼层
卡太老啦
回复

使用道具 举报

发表于 2022-3-10 15:36:36 | 显示全部楼层
的确,没想到GTX260还在干活
回复

使用道具 举报

发表于 2024-5-18 16:07:25 | 显示全部楼层
本帖最后由 zkzip 于 2024-5-29 10:13 编辑

嗯,我也基本只剩cpu了。以前好几块显卡都100%出错了。只剩下两块gtx660还能保一半。gfn19和以下基本全错,fgn20及以上基本能保正常。amd的hd6870也趴窝了。大多都在跑ap,能保过。
拿了几个系统测试了一下,win7需要装c++运行库,要不然所有任务秒错。装好之后所有任务都能跑,但显卡驱动过老的话很多任务都会跑个分钟后出错,有部分时间很长的包能正常。也就是说,只能装老版驱动的老显卡会很惨。
回复

使用道具 举报

发表于 2024-5-29 10:15:48 | 显示全部楼层
zflowers 发表于 2022-3-10 15:36
的确,没想到GTX260还在干活

别提了,我还有8800GTx都舍不得扔。740和640都算好东西了。260也有个能继续算的。
回复

使用道具 举报

发表于 2024-5-30 08:18:04 | 显示全部楼层
zkzip 发表于 2024-5-29 10:15
别提了,我还有8800GTx都舍不得扔。740和640都算好东西了。260也有个能继续算的。 ...

我记得GTX260发热量不低
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 新注册用户

本版积分规则

论坛官方淘宝店开业啦~

Archiver|手机版|小黑屋|中国分布式计算总站 ( 沪ICP备05042587号 )

GMT+8, 2024-10-13 20:47

Powered by Discuz! X3.5

© 2001-2024 Discuz! Team.

快速回复 返回顶部 返回列表