PathScale Open-Sources The EKOPath 4 Compiler Suite

[DeiF]

Massive improvement indeed. :O

Must...compile...whole system...with it...

[kacperpl1]

Must...compile...whole system...with it...

Just thought about the same.
Anyway where's the release? There's still nothing on pathscale's site...

[Michael]

Just thought about the same.
Anyway where's the release? There's still nothing on pathscale's site...

There was some miscommunication but I suspect it will be on their site shortly.

[oliw]

Just noticed this on the Pathscale CTO's twitter feed:

@michaellarabel - Sorry about the delays, but I'm extremely disappointed in your impatience to respect our delays for the release. --karma

Have to agree with him. Michael could have sat silently on this for all this time but instead dripped it out and essentially gave it away (smart move giving out the exact price) so that he could generate four posts of ad-juicing goodness. Pretty scummy.

Michael, if you want people to talk to you in confidence, you have to show you'll respect their secrecy when they explicitly ask for it.

[markg85]

Massive improvement indeed. :O

Must...compile...whole system...with it...

That might not be the best thing to do. Remember, the things benchmarked stop far are computational apps and those seem to benefit a lot from it. Buy we both don't have the slightest idea how this will perform on non computational apps like for example kde, gnome, firefox, office suits.. Besides that I'm also interested in the compiled binary size

More information is needed.
Note, I'm guessing firefox and chrome won't compile.

[Michael]

Have to agree with him. Michael could have sat silently on this for all this time but instead dripped it out and essentially gave it away (smart move giving out the exact price) so that he could generate four posts of ad-juicing goodness. Pretty scummy.

Michael, if you want people to talk to you in confidence, you have to show you'll respect their secrecy when they explicitly ask for it.

He was actually okay with the 'leaks', but we just had some miss communications at the last minute. Last I heard was Thursday/Friday/Monday and pinged several times via email on IRC on Friday, Saturday, Sunday, and this morning to see if anything had changed, and hadn't heard, so assumed everything was still on target.

[oliw]

He was actually okay with the 'leaks', but we just had some miss communications at the last minute. Last I heard was Thursday/Friday/Monday and pinged several times via email on IRC on Friday, Saturday, Sunday, and this morning to see if anything had changed, and hadn't heard, so assumed everything was still on target.

Then there's perhaps a lesson to be learnt about assumptions. Especially on something that you say has slipped numerous times already.

[markg85]

He was actually okay with the 'leaks', but we just had some miss communications at the last minute. Last I heard was Thursday/Friday/Monday and pinged several times via email on IRC on Friday, Saturday, Sunday, and this morning to see if anything had changed, and hadn't heard, so assumed everything was still on target.

Let me quote someone (don't know who buy the line is nice)
"assumption its the mother of all screw ups"

And I think you made one by posting the article. Not that I mind

[smitty3268]

I'm also interested in how this is done

Is it speculatively running code ahead of time on other cpu cores? What are the compile times and binary size like compared to GCC?

[puelapisse]

The Linux kernel can even be built with this high-performance compiler after applying a trivial patch.

where can we find this patch please?

also, I understand this compiler is Intel? 64 & AMD64 performance tuned but does anybody knows if it will run on 32bit?