And I have just downloaded the source code and will try to come up with a Linux version, however my mother just passed away so it may take me a few days to get to work on it.

Jim, I'm sorry to hear about your mother. You and your family have my prayers.

---

Thanks for your help Tetsuji!

I've got all necessary stuff (sources from Eric's links, Intel compiler + IPP + licenses), but I just can't get seti_boinc to configure. It keeps giving me this error (haven't seen that one before):

checking size of long int... configure: error: cannot compute sizeof (long int), 77
See `config.log' for more details.

config.log

configure:42105: checking size of long int
configure:42424: g++ -o conftest  -I/usr/include -I/usr/include/openssl -I/opt/intel/ipp/5.1/ia32/include -I/opt/intel/ipp/5.1/ia32/
tools/staticlib   -L/usr/lib -nodefaultlibs -L/opt/intel/ipp/5.1/ia32/lib conftest.cc  -lssl -lcrypto /usr/lib/libcrypto.a /usr/lib/
libssl.a -Wl,-Bdynamic -ldl -Wl,-Bdynamic -lm /usr/lib/libnsl.a -Wl,-Bdynamic -lrt /usr/lib/libz.a -Wl,-Bdynamic -lgcc_eh -Wl,-Bdyna
mic -lpthread -Wl,-Bdynamic -lc -lippcore -lippsmerged >&5
[color=red]/tmp/cc6OoaL7.o:(.eh_frame+0x11): undefined reference to `__gxx_personality_v0'[/color]
collect2: ld returned 1 exit status
configure:42427: $? = 1     
configure: program exited with status 1
configure: failed program was:
| /* confdefs.h.  */
|
| #define PACKAGE_NAME "setiathome_enhanced"
| #define PACKAGE_TARNAME "setiathome_enhanced"
| #define PACKAGE_VERSION "5.12"
| #define PACKAGE_STRING "setiathome_enhanced 5.12"

...and so on.

Anyone have an idea how to get it to go over that configure check? I'm pretty sure my libs are up to date. Will do an apt-get update/upgrade and try again then, but I'm not very hopeful about that.

Other than that, it compiled the BOINC client using the intel compiler fine.

I got the same messages when my "gcc -v" and "g++ -v" didn't match.
This was under an older Debian install.

Maybe downloading the correct g++ version will help.

Regards Hans

---

I've been looking at the Windows apps. -- basically because it sounds like it's just a matter of setting up the compiler and setting the right complier switches.

What I'm not sure about is where to get the sources.....

I found the CVS, but I'm not sure how it's organized.

---

Maybe it's not that hard to compile a Windows version after have done a Linux version first ?

No, it wouldn't be hard to do, it's just that we're getting back to the cost of the programs and libraries to do it with! I sure don't have the money to buy the compiler, even at the "student" rate! (and anyway, could you see me at my age going back to school! Haha!) There is (I think) a freeware Windows compiler for Linux (or it might be for Windows itself) however I don't think that it would do what is required to actually "optimize" the application through compiler flags. It may be worth looking into, but I seriously doubt getting as much out of it as a dedicated Windows/Intel compiler could.

I'm sorry to hear about your mom, Jim !

Thanks. It wasn't unexpected. She was 86 yrs. old and had been in a nursing home in bad health for the past two years, but thanks for everyone's sympathys.

---

Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.

Yeah sorry to hear about Your Mom passing Jim.

I'd try and optimize the app for Windows, But I have so little knowledge of C++ It isn't even funny. And of course the only compiler I have would be possibly a student version(I think) of Borland C++, I don't know which version although It was fairly recent and It does works in Windows XP.

---

CA HSR built a foundation, is laying Track!
PRR T1 Class 4-4-4-4 #5550 Loco, US's 1st HST

Sorry to hear about your loss Jim.








I am getting very frustrated with trying to build seti, I installed fedora core 4, got the latest updates, downloaded, compiled, and installed gcc-3.3.4.

I compiled the fftw libraries and installed them.

Boinc builds fine just like under suse, but! seti still gives me the same errors that I mention above.

I guess I really don't have a clue again!

---

Sorry to hear about your loss Jim.

Same from me Jim.

I am getting very frustrated with trying to build seti, I installed fedora core 4, got the latest updates, downloaded, compiled, and installed gcc-3.3.4.

I compiled the fftw libraries and installed them.

Boinc builds fine just like under suse, but! seti still gives me the same errors that I mention above.

I guess I really don't have a clue again!

I had even less luck with Centos.

A simple "up2date --install gcc" took longer than a complete debian installation :o)

Some libraries required by BOINC aren't provided by "up2date". You have to go hunting for them by yourself...

BOINC doesn't like the libcurl version that centos has, and I can't get a newer libcurl installed because it requires an _older_ version of openssl than what I have :o)

I guess that's enough of "RPM hell" for me.

I'll try building on an older Debian version next...

Regards Hans

---

I guess that's enough of "RPM hell" for me.

I'll try building on an older Debian version next...

Regards Hans

I'm not so sure now that it is a distro problem, I am going to stick with fedora for a while to see what happens, just because it is such a hassel to set it all up all over again, spent all day on setting it up and then trying repeatedly to get seti to compile.

---

quoting skywatch:
If you can't give him the respect due him, by following his request to STOP using his apps. Send him some cash to pay for the work he did and maybe he'll feel that we did appreciate his efforts.

Actually, I did make a donation - which is one reason I don't feel particularly compelled to stop using the applications. I do wonder, though, if affairs might have turned out somewhat differently had more others done the same.

quoting Crystallize:
He have no legal right to stop people who already using his apps. To continue to do so, it's not personal.

I agree with that, too. I was just pointing out that continuing to use Crunch3r's applications really has little to do with respect. I respect his ability, and I appreciate the effort he put into the work he did.

Frankly, it seems as if there are some people who will be unhappy as long as *any* optimized application exists and is being used. Since many of these people are the same ones that are heard saying things like "credit doesn't matter" and "it's all about the science", I really can't understand why it matters to them.

I mean, what's wrong with "you run your computer your way, and I'll run my computer my way", as long as we're both producing valid results?

Maybe it's not that hard to compile a Windows version after have done a Linux version first ?

No, it wouldn't be hard to do, it's just that we're getting back to the cost of the programs and libraries to do it with! I sure don't have the money to buy the compiler, even at the "student" rate! (and anyway, could you see me at my age going back to school! Haha!) There is (I think) a freeware Windows compiler for Linux (or it might be for Windows itself) however I don't think that it would do what is required to actually "optimize" the application through compiler flags. It may be worth looking into, but I seriously doubt getting as much out of it as a dedicated Windows/Intel compiler could.

I'm sorry to hear about your mom, Jim !

Thanks. It wasn't unexpected. She was 86 yrs. old and had been in a nursing home in bad health for the past two years, but thanks for everyone's sympathys.

May she rest in peace. You have my regards, Jim.

And as for the Windows version, yes, I do have all necessary things. Just haven't gotten them installed on my workstation yet - real life has a nasty habit of interfering.

So far, I've tackled the linux one since I had the compiler and lib installed there already. Not much luck yet, like most others who tried so far :) As to distributing these apps under my license, I'm not so sure that's going to be the best way.
Instead, I think I'd rather donate money (those 150 dollars for an edu license sound like a very reachable goal) to take the license out for S@H itself. That might benefit more, not sure if that fee is for all platforms or just linux. In any case, Windows Linux and OS X client efficiency should benefit from using that compiler. And yes, even AMD CPUs run noticably quicker using intel IPP code (contrary to what someone, I forget who, posted). This is due to IPP being optimized for SSE 1/2/3 (which modern AMD64 CPUs support fully). You can find loads of benchmark comparisons on Tom's Hardware or wherever.

However, having these licenses means I can at least compile stuff and release it to a test audience to try and see what's what. If there is really no source change required, maybe we can convince Eric to compile standard as well as K7/MMX/SSE/2/3 versions for PC. Obviously, this would be more work on releasing a new version.
However, as someone else suggested, using BOINC to find out what a specific host supports and them sending them a binary that's *already* optimized for their platform would really be the absolute pinnacle of user-friendly behaviour while at the same time yielding more scientific value.

And guess what, the playing field would be as level as possible for everyone.

As to the how of compiling all these instead of just one platform that runs everywhere - setting up a compile script and a cron job is something I'd gladly participate in. Shouldn't be too complicated.

Okay, that was more long-winded than planned. Anyway, back to playing with that source :)

---

Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information

quoting Crystallize:
He have no legal right to stop people who already using his apps. To continue to do so, it's not personal.

I agree with that, too. I was just pointing out that continuing to use Crunch3r's applications really has little to do with respect. I respect his ability, and I appreciate the effort he put into the work he did.

Frankly, it seems as if there are some people who will be unhappy as long as *any* optimized application exists and is being used. Since many of these people are the same ones that are heard saying things like "credit doesn't matter" and "it's all about the science", I really can't understand why it matters to them.

I mean, what's wrong with "you run your computer your way, and I'll run my computer my way", as long as we're both producing valid results?

Also If there weren't any optimized app, I'd swear the same ones would object to someting else, Like, Oh overclocked fast PCs or something along that line. As I think they just don't like someone being able to produce WUs faster than they can. In other words, plain old jealousy to Me. I await a new optimized app for the Windows XP/Vista OS and until one arrives that can replace Crunch3r's 5.12 SSE2 app, I'll keep using Crunch3r's 5.12 SSE2 app too. I too respect Crunch3r's ability and His work, But the Laws of Germany just don't apply in the USA So being flamed for not agreeing to that is just not right, Especially when I never flamed anybody on this issue, Some will blindly do what He commands, Just like some did just over 61 years ago and We all know Who won then and It wasn't the blind follwers, It was those Who said No.


Note: No flames are meant here in My statement.

---

CA HSR built a foundation, is laying Track!
PRR T1 Class 4-4-4-4 #5550 Loco, US's 1st HST

I too respect Crunch3r's ability and His work, But the Laws of Germany just don't apply in the USA So being flamed for not agreeing to that is just not right, Especially when I never flamed anybody on this issue, Some will blindly do what He commands, Just like some did just over 61 years ago and We all know Who won then and It wasn't the blind follwers, It was those Who said No.

Er. Please refrain from using the Holocaust (even if only as an allusion), Germany and the USA in one sentence. It might get someone quite upset (quite a bit more than already), no matter their ethnicity or nationality.

Polemic remarks like these (even if you didn't mean them that way) only serve to push us all further apart when we should really be thinking how to work together.

---

Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information

I too respect Crunch3r's ability and His work, But the Laws of Germany just don't apply in the USA So being flamed for not agreeing to that is just not right, Especially when I never flamed anybody on this issue, Some will blindly do what He commands, Just like some did just over 61 years ago and We all know Who won then and It wasn't the blind follwers, It was those Who said No.

Er. Please refrain from using the Holocaust (even if only as an allusion), Germany and the USA in one sentence. It might get someone quite upset (quite a bit more than already), no matter their ethnicity or nationality.

Polemic remarks like these (even if you didn't mean them that way) only serve to push us all further apart when we should really be thinking how to work together.

It may not have been the perfect wording, But still I wasn't referring to the holocaust, I was refering to the leader then in power as before He came to power He had a criminal past and as such wasn't suitable or right for the job.

At least I never said It didn't happen like some have, as I don't like that It did happen, As I could not have been born If someone had killed My dad at anzio in Italy as He only had a .45 cal 1911 colt pistol to defend Himself with(He was technically legally blind as He had 20/200 vision in one eye then), Nazis didn't like the disabled either and they killed them first.

Nor would I belittle such a thing as the holocaust either, as It was still murder. It's one thing for soldiers to shoot at other soldiers or insurgents(terrorists), It's something else to kill those Who can't fight back.

So It wasn't intended as a polemic remark as such. Besides even though this isn't the place to talk about It, It must not be allowed to be repeated or ignored.

That's It, Back to seti.

---

CA HSR built a foundation, is laying Track!
PRR T1 Class 4-4-4-4 #5550 Loco, US's 1st HST

Exactly, back to SETI. That was my original point, this being a community driven project without a tyrannical leader ;)

And thanks Hans, that seems to have been the problem - g++ and gcc versions not matching. Debian has a habit of not relinking each package's default..I updated both gcc and g++ at the same time with apt-get, one got changed, one did not.

Silly things :)

Now let's see if it will compile.

--edit
Actually (I'm using Sarge) it didn't upgrade g++ on apt-get upgrade because that would have installed a new libstdc++..sometimes Debian logic escapes me.

---

Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information

Exactly, back to SETI. That was my original point, this being a community driven project without a tyrannical leader ;)

And thanks Hans, that seems to have been the problem - g++ and gcc versions not matching. Debian has a habit of not relinking each package's default..I updated both gcc and g++ at the same time with apt-get, one got changed, one did not.

Silly things :)

Now let's see if it will compile.

--edit
Actually (I'm using Sarge) it didn't upgrade g++ on apt-get upgrade because that would have installed a new libstdc++..sometimes Debian logic escapes me.

Yeah computers can be silly and a bit aggravating at times, As I suffered through My share over the last 26 years.

---

CA HSR built a foundation, is laying Track!
PRR T1 Class 4-4-4-4 #5550 Loco, US's 1st HST

I've checked in a fix to the s_util.h problem, updated boinc and am rebuilding the tar files in the build directory. I haven't had the chance to try a linux build today, though.

Eric

---

@SETIEric@qoto.org (Mastodon)

hmmmm....still getting errors with the tarball from your site, the g++ no input files is gone, but it was replaced with :

make[2]: *** [seti_boinc-main.o] Error 1
make[2]: Leaving directory `/home/boog/build_boinc/seti_boinc/client'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/boog/build_boinc/seti_boinc'
make: *** [all] Error 2

the other one is still the same:

I'm hoping the s_util.h fix got rid of this one...

Don't worry too much abouyt the fakedata errors, unless you need to build fake workunits for testing purposes. For people who want to use SETI@home for their own data analysis (attention SETILeague), fakedata.cpp might be of interest. Then again, the splitter might be useful as well for such things.

Eric

make[2]: *** [fakedata-xml_util.o] Error 1

---

@SETIEric@qoto.org (Mastodon)

Hm,

I've tried both CVS and Eric's (older, still getting newer) archives. Even when I specify BOINCDIR=/path/to/boinc/sources, it still throws a warning (but it *is* including it, I updated BOINC from CVS because that's the only place boinc_worker_timer() is to be found), but I guess that's okay. However, now I'm getting different errors right after:

---------------
if icpc -DHAVE_CONFIG_H -I. -I. -I.. -march=pentium2 -m3dnow -mfpmath=387 -g -O2 -I/usr/include -I/usr/include/openssl
-I/opt/intel/ipp/5.1/ia32/include -I/opt/intel/ipp/5.1/ia32/tools/staticlib -fPIC -DPIC -DHAVE_CONFIG_H
-sox -xW -O3 -fp-model fast -no-prec-div -no-prec-sqrt -ipo -DHAVE_CONFIG_H -DTEXT_UI -DNDEBUG -DCLIENT
-I../db -I/root/boincstuff/boinc/api -I/root/boincstuff/boinc/lib -pthread -sox -xW -O3 -fp-model fast -no-prec-div -no-prec-sqrt -ipo -MT seti_boinc-analyzeFuncs.o -MD -MP -MF ".deps/seti_boinc-analyzeFuncs.Tpo" -c -o seti_boinc-analyzeFuncs.o `test -f 'analyzeFuncs.cpp' || echo './'`analyzeFuncs.cpp; \\
then mv -f ".deps/seti_boinc-analyzeFuncs.Tpo" ".deps/seti_boinc-analyzeFuncs.Po"; else rm -f ".deps/seti_boinc-analyzeFuncs.Tpo"; exit 1; fi
icpc: Command line warning: ignoring unknown option '-m3dnow'
icpc: Command line warning: ignoring unknown option '-mfpmath=387'
icpc: Command line warning: overriding '-marchpentium2' with '-xW'
icpc: Command line warning: overriding '-O2' with '-O3'
sincos.h(44): error: omission of exception specification is incompatible with previous function "sincos" (declared at line 82 of "/usr/include/bits/mathcalls.h")
void sincos(double angle, double *s, double *c);
^

sincos.h(45): error: omission of exception specification is incompatible with previous function "sincosf" (declared at line 82 of "/usr/include/bits/mathcalls.h")
void sincosf(float angle, float *s, float *c);
^

sincos.h(46): error: omission of exception specification is incompatible with previous function "sinf" (declared at line 66 of "/usr/include/bits/mathcalls.h")
float sinf(float angle);
^

sincos.h(47): error: omission of exception specification is incompatible with previous function "cosf" (declared at line 64 of "/usr/include/bits/mathcalls.h")
float cosf(float angle);
^

sincos.h(48): error: omission of exception specification is incompatible with previous function "atanf" (declared at line 59 of "/usr/include/bits/mathcalls.h")
float atanf(float angle);
^

/root/boincstuff/boinc/api/boinc_api.h(97): warning #64: declaration does not declare anything
extern void boinc_worker_timer();
^

/root/boincstuff/boinc/api/boinc_api.h(97): warning #1128: a storage class may not be specified here
extern void boinc_worker_timer();
^

-----IPP-----
-----mmx-----
../db/schema_master.h(31): warning #187: use of "=" where "==" may have been intended
return !(db_is_open=0);
^

---using ICC---
compilation aborted for analyzeFuncs.cpp (code 2)
make[2]: *** [seti_boinc-analyzeFuncs.o] Fehler 1
make[2]: Leaving directory `/root/boincstuff/seti_boinc.old/client'
make[1]: *** [all-recursive] Fehler 1
make[1]: Leaving directory `/root/boincstuff/seti_boinc.old'
make: *** [all] Fehler 2

----------------------

So the boinc_worker_timer is just a warning now, not an error.
Might be due to me tring to compile with ICC and --enable-3dnow (the compile host is an old Duron). So next I'll set it up on a VM on an AMD64 3500+ instead, see if that works out better. At least it'll be able to run most binaries to test as opposed to this host.

---

Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information

So I guess Debian Sarge isn't the platform of choice for compiling seti_boinc.
Neither is Slackware 10.1, by the way :)

Fedora Core is installing on a VM, let's see how that goes, since CentOS didn't seem to work for Hans either.

The linux client is proving tough to get to work, I'll have a whack at the Windows version today instead if FC doesn't change much.

---

Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information