All my units erroring out...

...on a GTS450. Here the log:

<core_client_version>6.10.56</core_client_version>
<![CDATA[
<message>
process exited with code 1 (0x1, -255)
</message>
<stderr_txt>
../../projects/milkyway.cs.rpi.edu_milkyway/milkyway_separation_0.82_x86_64-pc-linux-gnu__cuda_opencl: /lib/libc.so.6: version `GLIBC_2.11' not found (required by ../../projects/milkyway.cs.rpi.edu_milkyway/milkyway_separation_0.82_x86_64-pc-linux-gnu__cuda_opencl)

</stderr_txt>
]]>

Do i have a chance to get an app with hardcoded lib's inside or without that dependency?
I use driver version 270.26beta to solve the downclocking problem and the host crunches without hickups for dnetc, collatz and primegrid...

---

Come on, 47 views and no answers?

---

Just a thought, Did you upgrade the libc6 library? The upgrade from libc6_2.13-16 to libc6_2.13-17 changed the /lib64 symlink. If you did, try downgrading to the libc6_2.13-16 version and see what happens.

---

No, all my linux64-hosts have DotschUX/1.2 (based on Ubuntu8.04) installed.
An other problem is, i don't know how to compile and install a newer libc-version without screwing up my entire installation.

---

Maybe next release. GPU things aren't really old stuff friendly; going that far back for OpenCL is really pushing it.

Any news on this?

I made no step forward in the middletime. "cmake" is the last... The dependencies are a nightmare and instead of getting a simple Makefile, i get only error messages.

---

Any news on this?

I made no step forward in the middletime. "cmake" is the last... The dependencies are a nightmare and instead of getting a simple Makefile, i get only error messages.

The only external dependency right now should be OpenCL. What sort of errors?

Any news on this?

I made no step forward in the middletime. "cmake" is the last... The dependencies are a nightmare and instead of getting a simple Makefile, i get only error messages.

The only external dependency right now should be OpenCL. What sort of errors?

I have a Makefile now but compiling an app is something different:

boinc@vmware2k-3:~/Cuda/milkyway/milkywayathome_client$ make
CMake Error: Error in cmake code at
/home/boinc/Cuda/milkyway/milkywayathome_client/cmake_modules/StaticBOINC.cmake:34:
Parse error. Function missing ending ")". Instead found left paren with text "(".
CMake Error at CMakeLists.txt:47 (include):
include could not find load file:

/home/boinc/Cuda/milkyway/milkywayathome_client/cmake_modules/StaticBOINC.cmake


---------------------------------------------
Building: milkywayathome_client
System: Linux-2.6.27-17-generic
Build type: Release
Arch: x86_64
Install path: /usr/local
----
CMAKE version:
CMAKE binary: /usr/bin/cmake
CTEST binary: /usr/bin/ctest
CMAKE generator: Unix Makefiles
----
Project src dir: /home/boinc/Cuda/milkyway/milkywayathome_client
Project bin dir: /home/boinc/Cuda/milkyway/milkywayathome_client
Build tool: /usr/bin/make
C Compiler: /usr/bin/gcc
----
Double precision: ON
Separation: ON
N-Body: OFF
OpenCL separation: OFF
OpenCL nbody: OFF
BOINC application: ON
BOINC release: ON
----
---------------------------------------------
CMake Error: find_package could not find module FindOpenMP.cmake or a configuration file for package OpenMP.
Adjust CMAKE_MODULE_PATH to find FindOpenMP.cmake or set OpenMP_DIR to the directory containing a CMake configuration file for OpenMP. The file will have one of the following names:
OpenMPConfig.cmake
openmp-config.cmake

-- Found X11: /usr/lib/libX11.so
-- checking for modules 'QUIET;readline'
-- package 'QUIET' not found
-- package 'readline' not found
-- Could NOT find readline
-- Could NOT find Curses
-- Using BOINC from submodule
CMake Error: Error in cmake code at
/home/boinc/Cuda/milkyway/milkywayathome_client/cmake_modules/StaticBOINC.cmake:34:
Parse error. Function missing ending ")". Instead found left paren with text "(".
CMake Error at milkyway/CMakeLists.txt:27 (include):
include could not find load file:

/home/boinc/Cuda/milkyway/milkywayathome_client/cmake_modules/StaticBOINC.cmake


CMake Error: Error in cmake code at
/home/boinc/Cuda/milkyway/milkywayathome_client/separation/CMakeLists.txt:91:
Parse error. Function missing ending ")". Instead found left paren with text "(".
CMake Error: find_package could not find module FindGit.cmake or a configuration file for package Git.
Adjust CMAKE_MODULE_PATH to find FindGit.cmake or set Git_DIR to the directory containing a CMake configuration file for Git. The file will have one of the following names:
GitConfig.cmake
git-config.cmake

-- Configuring done
make: *** [cmake_check_build_system] Error 1

---

I have a Makefile now but compiling an app is something different:

What version are you trying to build?

http://milkyway.cs.rpi.edu/milkyway/download/src/milkyway_separation_0.88.tar.gz

---

I ran many n-body V.80 this morning along with milky@home 0.88's. ( all passed with flying colors ) This last download included n-body V.84 and more 0.88's. All the new n-body V.84 failed!!!!! I'd send all the pages of the reasons why but that's redundant because those that have preceaded me have the same problem. Is there anything I can do to solve this problem?

http://milkyway.cs.rpi.edu/milkyway/download/src/milkyway_separation_0.88.tar.gz

I updated the LinuxOpenCL binaries to 0.90 which should require an older glibc

I ran many n-body V.80 this morning along with milky@home 0.88's. ( all passed with flying colors ) This last download included n-body V.84 and more 0.88's. All the new n-body V.84 failed!!!!! I'd send all the pages of the reasons why but that's redundant because those that have preceaded me have the same problem. Is there anything I can do to solve this problem?

I have the same problem with Windows 7 x64 Home Premium on a 2nd gen. Intel Core i5 quad. Not only do all the nbody v.84s error out, but the longest initial time on them was 60 seconds. That's WAY shorter than they used to be.

These are the only out of the ordinary messages:

12/11/2011 10:33:37 AM | Milkyway@home | [error] State file error: missing application file milkyway_nbody_0.80_windows_x86_64__mt
12/11/2011 10:33:37 AM | Milkyway@home | [error] No app version found for app milkyway_nbody platform windows_x86_64 ver 80 classmt; discarding nbody-Plum_Embedded_185113_0

I have the same problem with Windows 7 x64 Home Premium on a 2nd gen. Intel Core i5 quad. Not only do all the nbody v.84s error out, but the longest initial time on them was 60 seconds. That's WAY shorter than they used to be.

These workunits should finish quickly. It looks like you aren't getting errors anymore.

Matt,

I have reset my linux boxes, but still get the following. Have the linux apps been updated??

../../projects/milkyway.cs.rpi.edu_milkyway/milkyway_nbody_0.84_x86_64-pc-linux-gnu__mt: /lib64/libc.so.6: version `GLIBC_2.7' not found (required by ../../projects/milkyway.cs.rpi.edu_milkyway/milkyway_nbody_0.84_x86_64-pc-linux-gnu__mt)

---

I have the same problem with Windows 7 x64 Home Premium on a 2nd gen. Intel Core i5 quad. Not only do all the nbody v.84s error out, but the longest initial time on them was 60 seconds. That's WAY shorter than they used to be.

These workunits should finish quickly. It looks like you aren't getting errors anymore.

No. No errors, Matt. Are these short WUs for testing purposes?

http://milkyway.cs.rpi.edu/milkyway/download/src/milkyway_separation_0.88.tar.gz

I updated the LinuxOpenCL binaries to 0.90 which should require an older glibc

Thanks, i downloaded one unit http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=33848906.

There is a second unit http://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=33857497 listed but the unit has not reached my host...


[edit]
Valid unit and got ~160 credits granted. :D

---

Matt,

I have reset my linux boxes, but still get the following. Have the linux apps been updated??

../../projects/milkyway.cs.rpi.edu_milkyway/milkyway_nbody_0.84_x86_64-pc-linux-gnu__mt: /lib64/libc.so.6: version `GLIBC_2.7' not found (required by ../../projects/milkyway.cs.rpi.edu_milkyway/milkyway_nbody_0.84_x86_64-pc-linux-gnu__mt)

Yes. I built them against a Ubuntu 8.04 chroot which requires Glibc 2.7 (from 2007). I had some issues trying to get something older working but I'll try to get something older working at some point.

No. No errors, Matt. Are these short WUs for testing purposes?

Not quite? They are trying a different model of how to select a dark matter shell around the dwarf model. Before we were using a smaller Plummer sphere inside of a larger dark one, but these try something different for selecting a dark halo.