Fantastical deep information about optimizing for Apple Silicon can now be found here: developer.apple.com/download/apple…

Example of why power draw is a poor proxy for how well a program is utilizing your hardware (i7-7700K CPU in this case).

In one case, execution port use is heavy and IPC is high, so the CPU's pipeline is running at high utilization. However power draw is low. Second case...

youtube told me this one might have limited visibility, so rts will be very appreciated

youtube.com/watch?v=3xottY…

New blog post: 'Entropy decoding in Oodle Data: x86-64 6-stream Huffman decoders' fgiesen.wordpress.com/2023/10/29/ent…

Here is Qualcomm #Oryon single threaded performance:

👉 VS Apple M2:
14% faster at 30% less power

👉 VS Intel i9-13980HK:
1% faster at 70% less power

oh my god

Who wouldn't want to read about the cursed ABI that lets you mix ARM64 code and X64 code in the same address space?

corsix.org/content/window…

I think the audience for corsix.org/content/window… is approximately the empty set right now, but that might change if this Arm64EC thing takes off.

I just published the first version of my illustrated SVE (ARMv9 SIMD) instruction list, with a truly absurd number of instruction diagrams and descriptions.

Check it out at dougallj.github.io/asil/

Quick first impressions of the new P-core in the A17 Pro:
mastodon.social/@dougall/11110…

s/FAT_MAGIC/__builtin_bswap32(FAT_CIGAM)/g

I work on Chrome and I'm looking at a bug where chrome.dll is unloaded when it shouldn't be. These crashes are 99.7% correlated with umppc*.dll - aka CrowdStrike. Is there someone who can look at this?

The mold linker 2.0.0 is out! Starting from this version, mold is available under the MIT license instead of AGPL. github.com/rui314/mold/re…

Please enjoy the cursed 'light mode' that I added to this page that already had a white background: dougallj.github.io/applecpu/fires… (link in the top right, depends on HDR, e.g. Chrome or Safari on a recent Apple device not in low power mode)

y-cruncher v0.8.1 is now ready. Speedups on of #AVX512 on #Zen4 grows from 2% to 20+%.

Download here: numberworld.org/y-cruncher/

twitter.com/Mysticial/stat…

If you're into bignums, The Hybrid NTT algorithm that was the backbone for years is now detailed here: numberworld.org/y-cruncher/int…

New blog post: propagating bounds through bitwise operations.

Already posted on that other social network, but this is for the twitter-only people

bitmath.blogspot.com/2023/07/propag…

Pro tip: unfollow people to make the most of your quota

if there is anything dear to you on this website:
- open your browser's devtools
- scroll over all the relevant tweets
- hit 'Save All As HAR'
- leave the processing to kind souls later