Thanks! Didn't notice that.
Besides the AVX support is there anything else new - is it worth updating on non-AVX CPUs?

Yes, the new library made it always at least "a little" (few percent) faster.

---

_\|/_
U r s

r411b (new!)

Thanks! Didn't notice that.
Besides the AVX support is there anything else new - is it worth updating on non-AVX CPUs?

---

If you were referring to me, that's what I was using/testing with. The r411 app for linux. I remember a long time ago I got a useful output, but I guess not anymore.

The linux executables of r411 or r411b (new!) are so called static executables. It should have everything needed build in. Therefor "ldd" will not show anymore library infos. "ldd" works only if the executable is of shared type.

---

_\|/_
U r s

running the opt ap on linux is a bit easier than what you are attempting

If you were referring to me, that's what I was using/testing with. The r411 app for linux. I remember a long time ago I got a useful output, but I guess not anymore.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

running the opt ap on linux is a bit easier than what you are attempting

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

I came across a report of late-onset Signal 11 at Einstein recently. Their stderr_txt had the decency to report

libgcc_s.so.1 must be installed for pthread_cancel to work

-- signal handler called

Maybe an installed library check - is that LDD? - would help.

LDD is supposed to inform you of what all it needs and what you have.

for example:

machine:~ # ldd /bin/vi
        linux-gate.so.1 =>  (0xffffe000)
        libm.so.6 => /lib/libm.so.6 (0xb77d8000)
        libncurses.so.5 => /lib/libncurses.so.5 (0xb779b000)
        libacl.so.1 => /lib/libacl.so.1 (0xb7792000)
        libc.so.6 => /lib/libc.so.6 (0xb7636000)
        libdl.so.2 => /lib/libdl.so.2 (0xb7631000)
        /lib/ld-linux.so.2 (0xb7815000)
        libattr.so.1 => /lib/libattr.so.1 (0xb762b000)
machine:~ #

However, when I throw a linux AP binary at it, I get "not a dynamic executable". I'm pretty sure it worked before. That may have been before libfft was included instead of referenced.

When I had the segment violation errors a while back, the output of LDD specifically mentioned that I needed GLIBC3.2 or higher and I only had I think 2.8? I don't remember the full details. I just remember that it did tell me what the problem was.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

I came across a report of late-onset Signal 11 at Einstein recently. Their stderr_txt had the decency to report

libgcc_s.so.1 must be installed for pthread_cancel to work

-- signal handler called

Maybe an installed library check - is that LDD? - would help.

Yeah, I don't think I've ever seen a Linux 5.06 stock app successfully run all the way through a WU. I do point at a glibc incompatibility though. When the opt app came out for ap_v505 (r168), my single core machine was running 64-bit linux at the time, and would instant-fail with a segment violation.

Somebody had mentioned that I should run the 'LDD' (in lower-case) command on the binary and look for problems. Everything was fine except GLIBC. I tried upgrading that, but since it was an older distro, 200+ packages and the kernel had to be updated, so I just opted to grab a then-recent version of the same distro. Problem fixed.

So that's usually my go-to when someone has a segment violation on a linux app (stock or optimized). Especially if it is within the first 60 seconds.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

that one is .... OUCH!

http://setiathome.berkeley.edu/workunit.php?wuid=905551594

more than 66 hrs of hard work (239,847.70 secs) and end with ??? ''Error while computing''
i am happy it s not me. lol

edit : oh ! that one is worst lol, same computer:

270,542.90 secs = 75 hrs of hard work ''Error while computing''
http://setiathome.berkeley.edu/workunit.php?wuid=909224268

Linux + stock AP app = no bueno

I had noticed a machine I was paired with a while back that was doing that. I was informed the linux 5.06 app has an issue.

Here is the post.

It is decisions like this that make me wonder what the project admins are smoking sometimes.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the BP6/VP6 User Group today!

that one is .... OUCH!

http://setiathome.berkeley.edu/workunit.php?wuid=905551594

more than 66 hrs of hard work (239,847.70 secs) and end with ??? ''Error while computing''
i am happy it s not me. lol

edit : oh ! that one is worst lol, same computer:

270,542.90 secs = 75 hrs of hard work ''Error while computing''
http://setiathome.berkeley.edu/workunit.php?wuid=909224268

Not surprised, looking at my wingmen I've only seen ONE Astropulse v505 v5.06 complete successfully over the last couple of months. Though they usually error out after a few seconds.

---

Member of the People Encouraging Niceness In Society club.

that one is .... OUCH!

http://setiathome.berkeley.edu/workunit.php?wuid=905551594

more than 66 hrs of hard work (239,847.70 secs) and end with ??? ''Error while computing''
i am happy it s not me. lol

edit : oh ! that one is worst lol, same computer:

270,542.90 secs = 75 hrs of hard work ''Error while computing''
http://setiathome.berkeley.edu/workunit.php?wuid=909224268

---

How long until this disappears?
http://setiathome.berkeley.edu/workunit.php?wuid=909492654

Check this one.

http://setiathome.berkeley.edu/workunit.php?wuid=631815542

It's from 2010.

Bad WU...BAD, Bad, bad! ;o)

---

I Desire Peace and Justice, Jim Scott (Mod-Ret.)

How long until this disappears?
http://setiathome.berkeley.edu/workunit.php?wuid=909492654

Check this one.

http://setiathome.berkeley.edu/workunit.php?wuid=631815542

It's from 2010.

---

How long until this disappears?
http://setiathome.berkeley.edu/workunit.php?wuid=909492654

Probably on the 5th of March, or when the last remaining cruncher that has it in cache has returned it.

---

"Learn from yesterday. Live for today. Hope for tomorrow." Albert Einstein
"With cats." kittyman

How long until this disappears?
http://setiathome.berkeley.edu/workunit.php?wuid=909492654

---

I Desire Peace and Justice, Jim Scott (Mod-Ret.)