Stories
Slash Boxes
Comments

SoylentNews is people

SoylentNews is powered by your submissions, so send in your scoop. Only 13 submissions in the queue.

Compiler Jump Threading

posted by martyb on Monday November 02 2015, @04:31AM   Printer-friendly
from the next-up:-jump-jiving dept.

Phoenix666 writes:

The "jump threading" compiler optimization (aka -fthread-jump) turns conditional into unconditional branches on certain paths at the expense of code size. For hardware with branch prediction, speculative execution, and prefetching, this can greatly improve performance. However, there is no scientific publication or documentation at all. The Wikipedia article is very short and incomplete.

The linked article has an illustrated treatment of common code structures and how these optimizations work.

Original Submission

 
This discussion has been archived. No new comments can be posted.
Compiler Jump Threading | Log In/Create an Account | Top | 34 comments | Search Discussion
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.

  • (Score: 4, Interesting) by pixeldyne on Monday November 02 2015, @08:34AM

    by pixeldyne (2637) on Monday November 02 2015, @08:34AM (#257410)

    The article that is, and the submission. This is the kind of content I'd like to see more on SN.

    Starting Score:    1  point
    Moderation   +2  
       Insightful=1, Interesting=1, Total=2
    Extra 'Interesting' Modifier   0  
    Karma-Bonus Modifier   +1  
    Total Score:   4  

  • (Score: 2) by Phoenix666 on Monday November 02 2015, @12:20PM

    by Phoenix666 (552) on Monday November 02 2015, @12:20PM (#257447) Journal

    Do you have more sources for this kind of article? Not many like this turn up in sources I read or the RSS log.

    --
    Washington DC delenda est.

  • (Score: 2) by TheRaven on Monday November 02 2015, @12:46PM

    by TheRaven (270) on Monday November 02 2015, @12:46PM (#257456) Journal
    I found the treatment quite superficial, though I agree that this topic would be quite interesting. It also read as an advert for their C compiler (which is C only, GPLv2, and therefore unlikely to be of significant interest to anyone).
    --
    sudo mod me up

    • (Score: 2) by Wootery on Monday November 02 2015, @03:30PM

      by Wootery (2341) on Monday November 02 2015, @03:30PM (#257530)

      Also, their compiler project seems totally pointless.

      Want a C compiler that generates an SSA intermediate-representation and does lots of optimisations? We have that, it's called Clang, it's awesome, and it's used everywhere such a compiler is needed.

      • (Score: 0) by Anonymous Coward on Thursday November 05 2015, @05:19PM

        by Anonymous Coward on Thursday November 05 2015, @05:19PM (#258942)

        gcc generates an SSA intermediate-representation and does lots of optimisations. It's used everywhere.

  • (Score: 2) by ledow on Monday November 02 2015, @01:23PM

    by ledow (5567) on Monday November 02 2015, @01:23PM (#257471) Homepage

    Topic yes.

    Treatment of it, no.

    I actually didn't find the explanation all that clear and I'm not a stranger to the way modern processors operate.

    The point of it also was clear - why would you need or want to do this? It didn't seems covered particularly well.

    I'd like more of this, but... literally.. more of this. The article was quite bare, I felt.

    • (Score: 3, Touché) by Phoenix666 on Monday November 02 2015, @04:04PM

      by Phoenix666 (552) on Monday November 02 2015, @04:04PM (#257542) Journal

      If you have better sources, please submit them. Many of us would like to see them. I don't see much like this.

      --
      Washington DC delenda est.