Live data from Hacker News

Viewing profile — mpu

mpu

HN member
Joined
Tue, Feb 19, 2013, 6:43 PM UTC
HN karma
318
Public activity
79 items

About mpu

No profile information was provided.

Recent public activity

  1. comment
    Comment #44006178

    We just released a tiny (~3kloc) Python library that implements state-of-the-art inference algorithms on GPU and provides performance similar to vLLM. We believe it's a great learn…

  2. story
  3. story
  4. story
  5. comment
    Comment #11559570

    Hi, thanks for the information, but LLVM still does not provide ABI compatibility. If you reduce the struct to 3 i32, it is passed in edi, esi, and edx on my machine. However accor…

  6. comment
    Comment #11559350

    I actually decided to take more time to answer your comment more throughly than others. Also, I TA'd twice the class from where you linked the article below, so I know about it :).…

  7. comment
    Comment #11556717

    It's only a goal I set to myself, if we can do better, heck let's do it! Keeping the code short, on the other hand, is really something I care about.

  8. comment
    Comment #11556493

    I think the extra load/stores clutter the IL. Also, QBE does not really "alternate" SSA/non-SSA, SSA form is built once at the beginning of the compilation pipeline and preserved l…

  9. comment
    Comment #11556400

    Cool, thank you guys!

  10. comment
    Comment #11556396

    At least, I can try! And also, we are seeing more and more certified C programs: see the DeepSpec NSF expedition grant, the Verified Software Toolchain, and the CertiKOS project fo…

  11. comment
    Comment #11556381

    I take this as a compliment. It means my design choice was totally valid, and maybe even a good one.

  12. comment
    Comment #11556375

    Then maybe I did not express myself in the best terms. QBE definitely supports stack slots and their registerization! Minic, a small C frontend shipped with QBE makes use of them. …

  13. comment
    Comment #11556074

    It's an alternative if you fit in the use case. I did not try to clone LLVM.

  14. comment
    Comment #11556045

    It's much much smaller (I think libfirm is over 100kloc, QBE is about 6k). But the major difference is the IL: I use a human-readable and easily-printable text IL. This means that …

  15. comment
    Comment #11555922

    Thank you for your words. It is often called NIH, but eh, I learned a lot! And I think that I made some modest improvements over LLVM, you can check them out in my comparison at ht…

  16. story
  17. comment
    Comment #11306529

    12,000 links is probably too much.

  18. story
  19. story
  20. story
  21. comment
    Comment #11080980

    Hurray, no floating points!

  22. comment
    Comment #11077113

    So you rewrote the whole thing.

  23. comment
    Comment #10864903

    You're wrong about the arrays, the standard guarantees that zeroing the bytes of an object of integer type will give a 0 value to the integer. Pointers are a different matter thoug…

  24. comment
    Comment #10864885

    Bullshit, on x64 all 32 or 64 bit base arithmetic operations have the exact same cost, except the 64 bit version is one byte longer because of a REX prefix. This is why 32 bits is …

  25. comment
    Comment #10569483

    Do you realize you do exactly what you say shouldn't be done? You just defined "a good place to start" by negation. Idiot.