Live data from Hacker News

Viewing profile — vsrinivas

vsrinivas

HN member
Joined
Thu, Nov 11, 2010, 12:55 AM UTC
HN karma
21
Public activity
20 items

About vsrinivas

https://ops101.org

Recent public activity

  1. comment
    Comment #49131201

    QEMU implementation of the backend of AppleParavirtGPU.kext, accelerated via a metal to vulkan translation layer.

  2. story
  3. comment
    Comment #48086494

    Emulator of the Voyager Flight Data Subsystem, based on very hard-earned docs

  4. story
  5. comment
    Comment #47746964

    Hey - I use the same, w/ both gemma4 and gpt-oss-*; some things I have to do for a good experience: 1) Pin to an earlier version of codex (sorry) - 0.55 is the best experience IME,…

  6. comment
    Comment #39085378

    FYI, Lumon HQ / the Eeron Saarinen masterpiece is Holmdel Bell Labs; this article is about the Murray Hill building.

  7. comment
    Comment #35715927

    Don't forget BECM on Gen2

  8. comment
    Comment #30768966

    I'm sorry to hear Jakov passed away - he was a brilliant, kind engineer.

  9. comment
    Comment #29312147

    Intel Cooper Lake - it's a 14 nm server part that fit a particular niche (4S support, VNNI/bfloat16). It was partly due to FB, but also partly to provide a platform that was larger…

  10. comment
    Comment #20650416

    The TLB bug (Errata 298, doc 41322 if you really care - while the processor was attempting to set the A/D bits in a page table entry, an L2->L3 eviction of that PTE could occur) wa…

  11. comment
    Comment #16080228

    Control Flow Enforcement (ENDBRANCH requirement at branch targets) looks like a nice feature, looking forward to it.

  12. comment
    Comment #15093586

    From the perspective of - 'take a fresh look at something we take for granted' - "A Preliminary Architecture for a Basic Data-Flow Processor" (Dennis & Misunas 1975) Focusing on th…

  13. comment
    Comment #12526120

    You mean LRO instead of GRO, right?

  14. comment
    Comment #12263480

    Just wondering, why do you think so? (It wasn't obvious to me why, given a first read of the docs; the fact that applications run in VMs in Flex seems more like an implementation d…

  15. comment
    Comment #12041181

    Yes it is. Recommend browsing for great UNIX / TCP history.

  16. comment
    Comment #4800120

    And DragonFly's dntpd! It's time sync algorithms are actually pretty cool.

  17. comment
    Comment #4740482

    That these benchmarks were done with 9.3-devel was important; part of what triggered this work was PostGres moving to a different way of mapping it's shared memory segment (mmap vs…

  18. comment
    Comment #4740474

    Still hard to say. In the network stack, I think the DragonFly approach is definitely better (than fine-grained data structure locking). In DF, connections are routed to per-CPU ne…

  19. comment
    Comment #3107895

    The first edition has opening brackets for functions on a new line. It also used a space before parenthesis for while loops and conditionals. Return types, when required, were on t…

  20. comment
    Comment #2547081

    Hi, If you're interested in reading a paper about the execution model, Jeffrey Hsu's 2004 "The DragonFlyBSD Operating System" ( http://www.dragonflybsd.org/presentations/dragonflyb…