{
  "root_post_id": "2060191401883619479",
  "posts": [
    {
      "post_id": "2060186437820010850",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "Took perhaps a 5 month vacation from at home programming, but getting back in the grove with 1hr/day.  Polishing up my WIN32 and Linux system call interfaces. Perhaps working towards a demo of GPU-side audio and network processing.",
      "timestamp": "2026-05-29 02:28:17",
      "media_urls": [],
      "reply_to_id": null,
      "quote_of_id": null,
      "metrics": {
        "reply_count": 3,
        "repost_count": 1,
        "like_count": 49,
        "view_count": 3893
      }
    },
    {
      "post_id": "2060189329801003343",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "Today's topic was Sleep [crap] vs NtDelayExecution [good], does seem useful to be able to NtAlertThread while a thread is sleeping (an analog to signaling while nanosleeping on Linux).",
      "timestamp": "2026-05-29 02:39:47",
      "media_urls": [],
      "reply_to_id": "2060186437820010850",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 1,
        "repost_count": 0,
        "like_count": 11,
        "view_count": 707
      }
    },
    {
      "post_id": "2060190230838223346",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "Futex users, there is the WakeByAddressAll and WaitOnAddress in Windows. But why use those when you can just directly call https://ntdoc.m417z.com/rtlwaitonaddress - but haven't tried that yet, and are the docs correct, why PLARGE_INTEGER for ms, when NtDelayExecution does the 100ns unit thing?",
      "timestamp": "2026-05-29 02:43:22",
      "media_urls": [],
      "reply_to_id": "2060189329801003343",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 2,
        "repost_count": 0,
        "like_count": 9,
        "view_count": 654
      }
    },
    {
      "post_id": "2060191401883619479",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "https://trickybitsblog.github.io/2024/02/25/timestamps.html is a nice related read, also I'm a little late to the https://ntdoc.m417z.com/system_hypervisor_user_shared_data party, but will have that sorted soon as well",
      "timestamp": "2026-05-29 02:48:01",
      "media_urls": [],
      "reply_to_id": "2060190230838223346",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 1,
        "repost_count": 0,
        "like_count": 5,
        "view_count": 242
      }
    },
    {
      "post_id": "2060192645582905577",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "Other thought, I typically mmap an extra LOG file, and write via atomic to grab a fixed width line. No stdout. The log wraps around and exists for multiple executions or even parallel processes. It's the way to do it ... BUT ...",
      "timestamp": "2026-05-29 02:52:57",
      "media_urls": [],
      "reply_to_id": "2060191401883619479",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 1,
        "repost_count": 0,
        "like_count": 4,
        "view_count": 210
      }
    },
    {
      "post_id": "2060193385411969118",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "... I think I'm going to just move the LOG file into the first N KiB of the CART file. So mmap'ed on CPU and GPU, and I can just bank on how \"less\" doesn't actually load the entire file while printing, so I can still see output on a console if needed for debug.",
      "timestamp": "2026-05-29 02:55:54",
      "media_urls": [],
      "reply_to_id": "2060192645582905577",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 1,
        "repost_count": 0,
        "like_count": 4,
        "view_count": 212
      }
    },
    {
      "post_id": "2060195139696365679",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "And another random thought, can you force de-power the CPU by abusing the https://www.felixcloutier.com/x86/pause instruction for things that are not spin-loops :) Going to have to try this one at some point.",
      "timestamp": "2026-05-29 03:02:52",
      "media_urls": [],
      "reply_to_id": "2060193385411969118",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 1,
        "repost_count": 0,
        "like_count": 4,
        "view_count": 183
      }
    },
    {
      "post_id": "2060195914568851775",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "Pushing data to GPU is really a no-brainer (write-combined stores), minus possibility of non-atomic write visibility, so I do simple stuff like always having a ring buffer of packets available where the GPU can validate each entry with a HASH, taking the latest valid one to use",
      "timestamp": "2026-05-29 03:05:57",
      "media_urls": [],
      "reply_to_id": "2060195139696365679",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 1,
        "repost_count": 0,
        "like_count": 5,
        "view_count": 176
      }
    },
    {
      "post_id": "2060196816574882209",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "Also AMD exports non-cached memory in VK, so it's possible to poll on data the CPU pushes (via write combined stores to VRAM) on AMD a few times per frame even.",
      "timestamp": "2026-05-29 03:09:32",
      "media_urls": [],
      "reply_to_id": "2060195914568851775",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 1,
        "repost_count": 0,
        "like_count": 3,
        "view_count": 182
      }
    },
    {
      "post_id": "2060197138923938169",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "The workaround for NVIDIA is if one wants to read N times per frame, then the CPU needs to duplicate the packet ring to N separate lines. Then the GPU is free to poll, but using separate cachelines each time.",
      "timestamp": "2026-05-29 03:10:49",
      "media_urls": [],
      "reply_to_id": "2060196816574882209",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 1,
        "repost_count": 0,
        "like_count": 4,
        "view_count": 239
      }
    },
    {
      "post_id": "2060198133905199300",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "The route back for IO that needs to be routed through the CPU {audio out, network out} is a lot more unfun. AMD again has the nice uncached memory support, so that part is easy. NVIDIA well, would have to take the crappy mid-frame L2 writeback (likely) making CPU-read available",
      "timestamp": "2026-05-29 03:14:46",
      "media_urls": [],
      "reply_to_id": "2060197138923938169",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 2,
        "repost_count": 0,
        "like_count": 4,
        "view_count": 570
      }
    },
    {
      "post_id": "2060199422730871174",
      "author": "NOTimothyLottes",
      "handle": "NOTimothyLottes",
      "text": "Those who have tread here before with driver side logic (DMA transfers/etc) got screwed, Win+Linux driver idiots and their CPU interrupt based scheduling = completely useless. So getting the driver out of the picture is the way to go",
      "timestamp": "2026-05-29 03:19:53",
      "media_urls": [],
      "reply_to_id": "2060198133905199300",
      "quote_of_id": null,
      "metrics": {
        "reply_count": 0,
        "repost_count": 0,
        "like_count": 4,
        "view_count": 508
      }
    }
  ],
  "source_url": "https://x.com/NOTimothyLottes/status/2060191401883619479"
}