Live data from Hacker News

Viewing profile — drv

drv

HN member
Joined
Wed, Jul 28, 2010, 6:01 PM UTC
HN karma
807
Public activity
171 items

About drv

daniel@drv.nu

Recent public activity

  1. comment
    Comment #40509485

    I/O scheduler was probably a bad example, since you might not need/want one for fast NVMe devices anyway, but yes, they help ensure limited resources (storage device bandwidth or I…

  2. comment
    Comment #40508367

    It's cool to see more systems code written in Rust! I also previously worked on SPDK, so it was neat to see it being chosen as a point of comparison. However, I was waiting for the…

  3. comment
    Comment #15126432

    The assembly version is using the packed version of the FMA instruction (that's what the "P" in the mnemonic stands for), but as far as I can tell, it's only using one of the packe…

  4. comment
    Comment #13769399

    nextAfter is probably also including the denormals (an additional 2^23 values near 0).

  5. comment
    Comment #13187874

    One possible reason is that storing a bool separately from the index makes it difficult to update the producer or consumer index atomically. With the implementations that store the…

  6. comment
    Comment #12980968

    It's presumably not standard Host Memory Buffer, since the spec says "The controller shall function properly without host memory resources."

  7. comment
    Comment #12842302

    2001, not 2011. Time flies. :)

  8. comment
    Comment #12756140

    (Assuming you mean MailChimp) I don't know anything about how MailChimp operates, but a quick search turns up this blog post about how to set up SPF records [1]. From there, you ca…

  9. story
  10. story
  11. comment
    Comment #11491468

    It has a C API, but everything under the hood is C++. The source is actually public now: https://github.com/Microsoft/Windows-Driver-Frameworks

  12. comment
    Comment #11491446

    Microsoft has been shipping static analysis tools with the Windows DDK for a long time (originally PREfast[1], now Static Driver Verifier[2]). I believe the static analysis is even…

  13. comment
    Comment #11012548

    In practice, this is probably correct (Microsoft cares a lot about backward compatibility, and many programs depend on MSVCRT.DLL). However, the official word from Microsoft is tha…

  14. comment
    Comment #10898287

    Anyone running FFmpeg[1] on untrusted input without sandboxing of some kind is being extremely negligent. It's around a million lines of C that does tricky file format parsing and …

  15. comment
    Comment #10518122

    Ah, I see. The I/OAT DMA copy offload is essentially equivalent to an asynchronous memcpy(), so anything addressable on the memory bus could be a source or destination (with some c…

  16. comment
    Comment #10517596

    The SPDK libraries are mostly storage-specific components (the I/OAT DMA engine can be used for generic copy offload, but it is particularly useful for copying between network and …

  17. comment
    Comment #10517189

    The NVMe driver will only work for a fairly narrow set of uses in which the whole NVMe device(s) can be dedicated to a single application (this is because the user-space applicatio…

  18. comment
    Comment #10517166

    I don't know anything about Omni-Path, sorry. However, based on the publicly available information, it does look like a very interesting combination. One major advantage of SPDK ov…

  19. comment
    Comment #10517036

    I am an engineer working at Intel on SPDK, and I can answer any technical questions you might have. Currently SPDK consists of a usermode NVMe (PCIe-attached SSD) driver. We will s…

  20. comment
    Comment #10516536

    It's certainly true that the polled-mode driver model doesn't interact well when the application needs to use other APIs that don't provide a polled mode. However, SPDK can be used…

  21. comment
    Comment #10199452

    This looks like an ideal case for something like ldexp(), but I couldn't find an equivalent function in Java at first glance. A lookup table seems like a bit of a shame for somethi…

  22. comment
    Comment #9866569

    The bufferbloat-related changes made in CeroWRT ( http://www.bufferbloat.net/projects/cerowrt ) have been merged into the Linux kernel and the OpenWrt project.

  23. comment
    Comment #9808060

    That's odd; the certificate I'm seeing for news.ycombinator.com has a SHA-256 signature, and the Qualsys test agrees.

  24. comment
    Comment #9535325

    The 55 MPH limit in the US wasn't even instituted for safety reasons; it was a reaction to the oil crisis. Any safety impact was probably due more to the improvements in car safety…

  25. story