Advanced search

Message boards : Graphics cards (GPUs) : unknown error_code

Author Message
Profile Sandro
Send message
Joined: 19 Aug 08
Posts: 22
Credit: 3,660,304
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 2081 - Posted: 4 Sep 2008 | 11:47:45 UTC

Hi,
a member of our team has this kind of error

<core_client_version>6.3.10</core_client_version>
<![CDATA[
<message>
Unzul�ssige Funktion. (0x1) - exit code 1 (0x1)
</message>
<stderr_txt>
# Using CUDA device 0
# Device 0: "GeForce 9600 GT"
# Clock rate: 1950000 kilohertz
MDIO ERROR: cannot open file "restart.coor"
Cuda error: Kernel [frc_sum_kernel_dihed] failed in file 'force.cu' in line 516 : unspecified launch failure.

</stderr_txt>
]]>

he is using WinXP32, boinc-version 6.3.10 and the 177.93 driver.
My idea was that his shader clock-rate is to high? or is something else wrong?
Sandro

Profile MJH
Project administrator
Project developer
Project scientist
Send message
Joined: 12 Nov 07
Posts: 696
Credit: 27,266,655
RAC: 0
Level
Val
Scientific publications
watwat
Message 2082 - Posted: 4 Sep 2008 | 11:55:49 UTC - in response to Message 2081.


# Device 0: "GeForce 9600 GT"
# Clock rate: 1950000 kilohertz
Cuda error: Kernel [frc_sum_kernel_dihed] failed in file 'force.cu' in line 516 : unspecified launch failure.


This is almost certainly due to instability caused by the high clock rate. The reference 9600GT design specifies a clock of 1625MHz. This user should use the NVidia control panel to reduce the shader clock.

MJH

Profile Sandro
Send message
Joined: 19 Aug 08
Posts: 22
Credit: 3,660,304
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 2083 - Posted: 4 Sep 2008 | 11:58:13 UTC - in response to Message 2082.


# Device 0: "GeForce 9600 GT"
# Clock rate: 1950000 kilohertz
Cuda error: Kernel [frc_sum_kernel_dihed] failed in file 'force.cu' in line 516 : unspecified launch failure.


This is almost certainly due to instability caused by the high clock rate. The reference 9600GT design specifies a clock of 1625MHz. This user should use the NVidia control panel to reduce the shader clock.

MJH


THX, this was also my idea. I will tell him this.

Profile Sandro
Send message
Joined: 19 Aug 08
Posts: 22
Credit: 3,660,304
RAC: 0
Level
Ala
Scientific publications
watwatwatwat
Message 2084 - Posted: 4 Sep 2008 | 12:37:51 UTC - in response to Message 2083.

another question ;)
i had one WU some days before who crashed with this error


<core_client_version>6.3.8</core_client_version>
<![CDATA[
<message>
process exited with code 193 (0xc1, -63)
</message>
<stderr_txt>
# Using CUDA device 0
# Device 0: "GeForce 8800 GT"
# Clock rate: 1674000 kilohertz
SIGSEGV: segmentation violation
Stack trace (8 frames):
acemd_6.42_x86_64-pc-linux-gnu__cuda[0x4b3bf9]
/lib/libc.so.6[0x7fbab0182100]
/lib/libc.so.6(fclose+0x7)[0x7fbab01b2c67]
acemd_6.42_x86_64-pc-linux-gnu__cuda[0x40fdc7]
acemd_6.42_x86_64-pc-linux-gnu__cuda(sin+0x1649)[0x4085a9]
acemd_6.42_x86_64-pc-linux-gnu__cuda(sin+0x33b)[0x40729b]
/lib/libc.so.6(__libc_start_main+0xf4)[0x7fbab016e1c4]
acemd_6.42_x86_64-pc-linux-gnu__cuda(sinh+0x59)[0x407079]

Exiting...

</stderr_txt>
]]>

running under ubuntu 8.04 64bit. all other WU are fine.
what was the error and how can i avoid it in future?
thx

Post to thread

Message boards : Graphics cards (GPUs) : unknown error_code

//