And sorry for my mistake "Ligo with GPL". No. Ligo with NDA or an access password or with abrakadabra bla!. (I can not translate that to english. - It has a meaning of hocus pocus, snake oil, ...)
In the meantime.. HAs any of you played with the LD_PRELOAD and pthread_cond_wait - and saved the original call address and done a usleep(0, 1) before or after that call.
Replacing sched_yield does not not work with E@h albeit it did its job some 10 years ago with S@h astropulse OpenCL and on those other SW before I made the (THE) version that had a chance against those apps that day and then on after I hit the slowest parts plus some other ...
I have a feeling that I downloaded this branch too, but If I remember correctly it misses (I may be wromg) the einstein at home branch that (I need to get from somewhere else. (I'll check).
And sorry for my mistake "Ligo with GPL". No. Ligo with NDA or an access password or with abrakadabra bla!. (I can not translate that to english. - It has a meaning of hocus pocus, snake oil, ...)
The last time you foisted your childish rants on the volunteers, you were given quite detailed responses about why these sorts of 'performances' were a wasted effort and highly likely to be completely counter-productive for you.
I wrote you a suggestion that would have a rather better chance of success than what you are currently doing. Did you ever bother to read it?
Gary, the project is releasing apps under GPL and therefore have the obligation to provide the source code. It's fairly simple. no need for a long diatribe.
petri is more than capable to make improvements, and it's not just the GW app that needs it. improvements could certainly be made to the nvidia GRP app, especially if petri can work the same kind of magic that he did at SETI. maybe the SETI GPU devs thought their openCL app was as good as could be made, then petri came along and made it 4-5x faster under CUDA (after several iterations of course).
Gary, the project is releasing apps under GPL and therefore have the obligation to provide the source code. It's fairly simple. no need for a long diatribe.
I'm fairly certain that the latest GW GPU app is NOT currently GPL'd. I remember a post from Bernd a long time ago saying that some sort of permission from the LIGO community was needed before that could happen - or something along those lines. There has been nothing official since and, as has already been noted, any source code linked to is very old and probably not currently being used.
Because I don't know the current status for sure, and because none of the volunteers posting here will know for sure either, why doesn't PETRI33 send an email to those who can make an official statement about the matter rather than continuing to rant on here?
If PETRI33 (or anyone else) is certain the code in question is GPL'd and consequently being illegally withheld, send an email requesting their rights be honoured forthwith. That should clarify the matter very quickly.
this is not ranting. I'm willing to help and try as a volunteer. As anyone should have right to. Instead of
"If you believe you have the necessary skill-set to contribute to the development of the GW GPU app, then you should put in the time and effort to compile a fully documented list of your previous history and achievements and any pertinent information about what you are offering to do if given access to the codebase. It's a serious business so make a serious attempt to showcase what you are able to bring to the table. You should send those details and any supporting documentation to the general email address and marked to the attention of Bruce Allen. You should expect to go through a fairly exhaustive vetting procedure, including the signing of NDAs, before being given access to the code. "
And thanks. I was kicked off and back with that response. It is not so serious. It is a hobby and fun.
I reread it and found new u s e f u l information from the university web site you referred to. Your good message was hidden in a lot of ****.
The thread is still open, should someone know where to get the latest source code. It is not open for comment saying I'm ranting. I just need some help, some friendly help, to make my first steps in to the wonderful world of optimization as a hobby.
following :D
)
following :D
looks like you can get it from here: https://einsteinathome.org/application-source-code-and-license
_________________________________________________________________________
Hi Ian & Steve, Nice to
)
Hi Ian & Steve,
Nice to hear from You ...
.. But no no. Not at all. All links are outdated
No current source code can be found in two weeks of extensive search.
Some links point to Ligo with GPL and most of the links point to 2011 or before that.
I love that You answered to me.
So Who'd like to take the next try and guess?
And sorry for my mistake
)
And sorry for my mistake "Ligo with GPL". No. Ligo with NDA or an access password or with abrakadabra bla!. (I can not translate that to english. - It has a meaning of hocus pocus, snake oil, ...)
is what you're looking for
)
is what you're looking for not located here?: https://git.ligo.org/lscsoft/lalsuite/
looks like it has recent activity.
_________________________________________________________________________
.. and In the meantime.. HAs
)
.. and
In the meantime.. HAs any of you played with the LD_PRELOAD and pthread_cond_wait - and saved the original call address and done a usleep(0, 1) before or after that call.
Replacing sched_yield does not not work with E@h albeit it did its job some 10 years ago with S@h astropulse OpenCL and on those other SW before I made the (THE) version that had a chance against those apps that day and then on after I hit the slowest parts plus some other ...
Thanks, I have a feeling
)
Thanks,
I have a feeling that I downloaded this branch too, but If I remember correctly it misses (I may be wromg) the einstein at home branch that (I need to get from somewhere else. (I'll check).
Thanks,
I love You.
petri33 wrote:And sorry for
)
The last time you foisted your childish rants on the volunteers, you were given quite detailed responses about why these sorts of 'performances' were a wasted effort and highly likely to be completely counter-productive for you.
I wrote you a suggestion that would have a rather better chance of success than what you are currently doing. Did you ever bother to read it?
Cheers,
Gary.
Gary, the project is
)
Gary, the project is releasing apps under GPL and therefore have the obligation to provide the source code. It's fairly simple. no need for a long diatribe.
petri is more than capable to make improvements, and it's not just the GW app that needs it. improvements could certainly be made to the nvidia GRP app, especially if petri can work the same kind of magic that he did at SETI. maybe the SETI GPU devs thought their openCL app was as good as could be made, then petri came along and made it 4-5x faster under CUDA (after several iterations of course).
_________________________________________________________________________
Ian&Steve C. wrote:Gary, the
)
I'm fairly certain that the latest GW GPU app is NOT currently GPL'd. I remember a post from Bernd a long time ago saying that some sort of permission from the LIGO community was needed before that could happen - or something along those lines. There has been nothing official since and, as has already been noted, any source code linked to is very old and probably not currently being used.
Because I don't know the current status for sure, and because none of the volunteers posting here will know for sure either, why doesn't PETRI33 send an email to those who can make an official statement about the matter rather than continuing to rant on here?
If PETRI33 (or anyone else) is certain the code in question is GPL'd and consequently being illegally withheld, send an email requesting their rights be honoured forthwith. That should clarify the matter very quickly.
Cheers,
Gary.
Hi, this is not ranting. I'm
)
Hi,
this is not ranting. I'm willing to help and try as a volunteer. As anyone should have right to. Instead of
"If you believe you have the necessary skill-set to contribute to the development of the GW GPU app, then you should put in the time and effort to compile a fully documented list of your previous history and achievements and any pertinent information about what you are offering to do if given access to the codebase. It's a serious business so make a serious attempt to showcase what you are able to bring to the table. You should send those details and any supporting documentation to the general email address and marked to the attention of Bruce Allen. You should expect to go through a fairly exhaustive vetting procedure, including the signing of NDAs, before being given access to the code. "
And thanks. I was kicked off and back with that response. It is not so serious. It is a hobby and fun.
I reread it and found new u s e f u l information from the university web site you referred to. Your good message was hidden in a lot of ****.
The thread is still open, should someone know where to get the latest source code. It is not open for comment saying I'm ranting. I just need some help, some friendly help, to make my first steps in to the wonderful world of optimization as a hobby.