Seti application expieriences SSE3 emulation crashes on Jaunty Jackalope 9.04

[michael37]

Binary package hint: boinc-app-seti

I am reporting a regression in 64-bit Optimized Seti@Home applications in Jaunty. It appears that the problem affects Core 2 Duo 'Merom' laptops using SSE2->SSE3 emulation. The problem did not appear in Intrepid. Other similar OS'es, such as Sidux, sharing the same kernel, do not exhibit the problem.

Current workaround is (source: http://lunatics.kwsn.net/linux/64-bit-sse2-app.15.html)
"So, if you want to use the optimized application version astropulse 5.03 x64 sse3 with Ubuntu/Kubuntu/Xubuntu 9.04 and kernel 2.6.28-11-generic on a core 2 mobile with "Merom" core, please use astropulse 5.03 x64 sse2 optimized application version instead.
More testing has to be done to find a better solution for this problem."

Here are several stack traces (source: http://lunatics.kwsn.net/linux/more-trouble-with-x-k-ubuntu-9-04-64bit.0.html). It looks like the cause of the problem may be somewhere in libpthreads.

Name 07fe09aa.24575.21749.15.8.2_0
Workunit 438668091
Created 1 May 2009 23:06:02 UTC
Sent 2 May 2009 4:36:41 UTC
Received 3 May 2009 14:13:32 UTC
Server state Over
Outcome Client error
Client state Compute error
Exit status 131 (0x83)
Computer ID 4593585
Report deadline 25 May 2009 0:10:27 UTC
CPU time 2950.788412
stderr out

<core_client_version>6.4.5</core_client_version>
<![CDATA[
<message>
process exited with code 131 (0x83, -125)
</message>
<stderr_txt>
Linux optimized S@H Enhanced application by Alex Kan
Version info: SSSE3x (Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSSE3x Linux64 Build 46 PGO, Ported by : Jason G, Raistmer, JDWhale

Processor Information:
  Model: Intel Core 2 Quad Q6600, 2.97 GHz
  Package: 4 Cores

Processor Caches:
  L1 code cache, 32 KB
  L1 data cache, 32 KB
  L2 combined cache, 4 MB

Processor Features: 64bit simd [x86] cmov mmx sse sse2 sse3 ssse3 vmx lm lahf_lm tm tm2 eist nx

Work Unit Info:
...............
Credit multiplier is : 2.85
WU true angle range is : 0.448236
terminate called after throwing an instance of 'std::bad_alloc'
  what(): St9bad_alloc
SIGABRT: abort calledStack trace (18 frames):
AK_V8_linux64_ssse3(boinc_catch_signal+0x17d)[0x43b70d]
/lib/libpthread.so.0[0x7fa389ed6080]
/lib/libc.so.6(gsignal+0x35)[0x7fa389b87fb5]
/lib/libc.so.6(abort+0x183)[0x7fa389b89bc3]
AK_V8_linux64_ssse3[0x576234]
AK_V8_linux64_ssse3[0x56a706]
AK_V8_linux64_ssse3[0x56a733]
AK_V8_linux64_ssse3[0x56cafa]
AK_V8_linux64_ssse3[0x56c539]
AK_V8_linux64_ssse3[0x5cf0df]
AK_V8_linux64_ssse3[0x5bd1c5]
AK_V8_linux64_ssse3[0x419d0a]
AK_V8_linux64_ssse3[0x415950]
AK_V8_linux64_ssse3[0x4149a2]
AK_V8_linux64_ssse3[0x405f7a]
AK_V8_linux64_ssse3[0x405612]
/lib/libc.so.6(__libc_start_main+0xe6)[0x7fa389b735a6]
AK_V8_linux64_ssse3(realloc+0x181)[0x405379]

Exiting...

</stderr_txt>
]]>

Validate state Invalid
Claimed credit 38.7802458050881
Granted credit 0
application version 5.28

<core_client_version>6.4.5</core_client_version>
<![CDATA[
<message>
process exited with code 131 (0x83, -125)
</message>
<stderr_txt>
Linux optimized S@H Enhanced application by Alex Kan
Version info: SSE3 (AMD/Intel, Core 2-optimized v8-nographics) V5.13 by Alex Kan
SSE3 Linux64 Build 46 PGO, Ported by : Jason G, Raistmer, JDWhale

Processor Information:
  Model: Intel Core 2 T5300, 1.73 GHz
  Package: 2 Cores

Processor Caches:
  L1 code cache, 32 KB
  L1 data cache, 32 KB
  L2 combined cache, 2 MB

Processor Features: 64bit simd [x86] cmov mmx sse sse2 sse3 ssse3 lm lahf_lm tm tm2 eist nx

Work Unit Info:
...............
Credit multiplier is : 2.85
WU true angle range is : 0.415339
terminate called after throwing an instance of 'std::bad_alloc'
  what(): St9bad_alloc
SIGABRT: abort calledStack trace (18 frames):
AK_V8_linux64_sse3(boinc_catch_signal+0x17d)[0x4358ed]
/lib/libpthread.so.0[0x7f564ea6f080]
/lib/libc.so.6(gsignal+0x35)[0x7f564e720fb5]
/lib/libc.so.6(abort+0x183)[0x7f564e722bc3]
AK_V8_linux64_sse3[0x570c64]
AK_V8_linux64_sse3[0x565136]
AK_V8_linux64_sse3[0x565163]
AK_V8_linux64_sse3[0x56752a]
AK_V8_linux64_sse3[0x566f69]
AK_V8_linux64_sse3[0x590d47]
AK_V8_linux64_sse3[0x58e86b]
AK_V8_linux64_sse3[0x418fce]
AK_V8_linux64_sse3[0x4159cb]
AK_V8_linux64_sse3[0x414a14]
AK_V8_linux64_sse3[0x406074]
AK_V8_linux64_sse3[0x405612]
/lib/libc.so.6(__libc_start_main+0xe6)[0x7f564e70c5a6]
AK_V8_linux64_sse3(realloc+0x181)[0x405379]

Exiting...

</stderr_txt>
]]>

<core_client_version>6.4.5</core_client_version>
<![CDATA[
<message>
process exited with code 193 (0xc1, -63)
</message>
<stderr_txt>
In ap_gfx_main.cpp: in ap_graphics_init(): Starting client.
AstroPulse v. 5.03
Linux 64 bit
FFTW USE_CONVERSION_OPT USE_SSE2->SSE3_EMU
SIGSEGV: segmentation violation
Stack trace (6 frames):
astropulse-5.03.x86_64-pc-linux-gnu(boinc_catch_signal+0x43)[0x435b23]
/lib/libpthread.so.0[0x7fbcb9005080]
astropulse-5.03.x86_64-pc-linux-gnu[0x408739]
astropulse-5.03.x86_64-pc-linux-gnu[0x406537]
/lib/libc.so.6(__libc_start_main+0xe6)[0x7fbcb8ca25a6]
astropulse-5.03.x86_64-pc-linux-gnu(__gxx_personality_v0+0x199)[0x405ca9]

Exiting...

</stderr_txt>
]]>

[michael37 (misha37)]

The segmentation always occurs in /lib/libpthreads.so.0 on 64-bit (amd64).

[michael37 (misha37)]

A set of problems reported by an owner of another type CPU.

CPU type GenuineIntel Intel(R) Core(TM)2 Quad CPU Q9300 @ 2.50GHz [Family 6 Model 23 Stepping 7]
Number of processors 4
Coprocessors NVIDIA Quadro NVS 290 (255MB)
Operating System Linux 2.6.28-11-generic

SSE-3 application:

<core_client_version>6.2.15</core_client_version>
<![CDATA[
<message>
process exited with code 193 (0xc1, -63)
</message>
<stderr_txt>
In ap_gfx_main.cpp: in ap_graphics_init(): Starting client.
AstroPulse v. 5.03
Linux 64 bit
FFTW USE_CONVERSION_OPT USE_SSE3
In ap_gfx_main.cpp: in ap_graphics_init(): Starting client.
### Restart at 88.16 percent.
SIGSEGV: segmentation violation
Stack trace (6 frames):
astropulse-5.03.x86_64-pc-linux-gnu(boinc_catch_signal+0x43)[0x4325d3]
/lib/libpthread.so.0[0x7f3061613080]
astropulse-5.03.x86_64-pc-linux-gnu[0x408666]
astropulse-5.03.x86_64-pc-linux-gnu[0x406536]
/lib/libc.so.6(__libc_start_main+0xe6)[0x7f30612b05a6]
astropulse-5.03.x86_64-pc-linux-gnu(__gxx_personality_v0+0x199)[0x405ca9]

Exiting...

</stderr_txt>
]]>

SSE-2 application:

<core_client_version>6.6.20</core_client_version>
<![CDATA[
<message>
process exited with code 193 (0xc1, -63)
</message>
<stderr_txt>
In ap_gfx_main.cpp: in ap_graphics_init(): Starting client.
AstroPulse v. 5.03
Linux 64 bit
FFTW USE_CONVERSION_OPT USE_SSE2->SSE3_EMU
In ap_gfx_main.cpp: in ap_graphics_init(): Starting client.
### Restart at 7.51 percent.
SIGSEGV: segmentation violation
Stack trace (7 frames):
astropulse-5.03.x86_64-pc-linux-gnu(boinc_catch_signal+0x43)[0x435b23]
/lib/libpthread.so.0[0x7ff1c253f080]
astropulse-5.03.x86_64-pc-linux-gnu[0x409574]
astropulse-5.03.x86_64-pc-linux-gnu[0x406f7e]
astropulse-5.03.x86_64-pc-linux-gnu[0x406537]
/lib/libc.so.6(__libc_start_main+0xe6)[0x7ff1c21dc5a6]
astropulse-5.03.x86_64-pc-linux-gnu(__gxx_personality_v0+0x199)[0x405ca9]

Exiting...

</stderr_txt>
]]>

From http://lunatics.kwsn.net/linux/64-bit-sse2-app.msg17330.html#msg17330:

I, too, have started experiencing work unit errors with optimized astropulse when I upgraded to Jaunty. In my case almost every work unit failed (a few, say, ~10% finish successfully).

This is a on a Yorkfield core (Intel Core Quad Q9300).

Note: The optimized MB SSE3 and SSE2 (editor: other applications) suffer the same fate as optimized astropulse on my system.
Editor: the similarities are (1) all applications are build with intel compiler and (2) all applications tend to heavily stress CPUs.

Other possibly relevant information:
Einstein@Home is the only other project running on this machine
VMware Server 2.0.1 is used to run a Windows XP guest
This was rock solid under Ubuntu 8.10 (i.e. never experienced a single work unit error).
I tried both BOINC 6.2.15 and 6.6.20 to see if it made any difference, but it did not.

[michael37 (misha37)]

The problem appears to be extremely widespread and affects Jaunty only.

http://setiathome.berkeley.edu/forum_thread.php?id=53399#890776

Please keep in mind that this is not a single application, but a wide spectrum of different binaries.

[michael37 (misha37)]

Verified that the issue is not glibc -- it's actually in the kernel itself. Karmic with 2.6.31 kernel does not exhibit this problem. Karmic with Jaunty's 2.6.28 kernel does.