<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: my123</title><link>https://news.ycombinator.com/user?id=my123</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 17 Apr 2026 09:26:03 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=my123" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[Drawbridge: What SQL Server on Linux is built on (2021)]]></title><description><![CDATA[
<p>Article URL: <a href="https://threedots.ovh/blog/2021/01/drawbridge-what-sql-server-on-linux-is-built-on/">https://threedots.ovh/blog/2021/01/drawbridge-what-sql-server-on-linux-is-built-on/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47756116">https://news.ycombinator.com/item?id=47756116</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 13 Apr 2026 18:32:10 +0000</pubDate><link>https://threedots.ovh/blog/2021/01/drawbridge-what-sql-server-on-linux-is-built-on/</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47756116</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47756116</guid></item><item><title><![CDATA[New comment by my123 in "Investigating Split Locks on x86-64"]]></title><description><![CDATA[
<p>Redacted</p>
]]></description><pubDate>Sat, 11 Apr 2026 14:46:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=47731112</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47731112</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47731112</guid></item><item><title><![CDATA[New comment by my123 in "Investigating Split Locks on x86-64"]]></title><description><![CDATA[
<p>XNU has this oddity: <a href="https://github.com/apple-oss-distributions/xnu/blob/f6217f891ac0bb64f3d375211650a4c1ff8ca1ea/osfmk/arm64/sleh.c#L1756" rel="nofollow">https://github.com/apple-oss-distributions/xnu/blob/f6217f89...</a><p>Redacted from open source XNU, but exists in the closed source version</p>
]]></description><pubDate>Sat, 11 Apr 2026 06:25:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=47728002</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47728002</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47728002</guid></item><item><title><![CDATA[New comment by my123 in "IBM Announces Strategic Collaboration with Arm"]]></title><description><![CDATA[
<p>After Power9, IBM became uncompetitive multi-core performance against mainstream server CPUs - both x86 and Arm. They didn't keep up with the rise in core counts.<p>And the single thread side isn't that good either, but SMT8 is a quite nice software licensing trick</p>
]]></description><pubDate>Thu, 02 Apr 2026 15:29:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=47615794</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47615794</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47615794</guid></item><item><title><![CDATA[New comment by my123 in "People inside Microsoft are fighting to drop mandatory Microsoft Account"]]></title><description><![CDATA[
<p>Xcode needs an Apple ID for download but the macOS SDK and toolchain does not.<p>Try to run any developer tool or "xcode-select install" and it'll download the command-line tools independently from Xcode.<p>(and then bring your own IDE)</p>
]]></description><pubDate>Fri, 27 Mar 2026 17:46:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=47545908</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47545908</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47545908</guid></item><item><title><![CDATA[New comment by my123 in "What happened to GEM?"]]></title><description><![CDATA[
<p>There were two entirely separate DOS 4s<p>The multi-tasking DOS 4 that had its next release be branded as OS/2<p>and the other one, a conventional DOS stopgap product</p>
]]></description><pubDate>Thu, 26 Mar 2026 01:03:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=47525503</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47525503</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47525503</guid></item><item><title><![CDATA[New comment by my123 in "Fedora 44 on the Raspberry Pi 5"]]></title><description><![CDATA[
<p>> Which vendors already support this?<p>One implementation I've seen in the wild is: <a href="https://docs.nvidia.com/jetson/archives/r36.4/DeveloperGuide/SD/Bootloader/UEFI.html#miniuefi-support" rel="nofollow">https://docs.nvidia.com/jetson/archives/r36.4/DeveloperGuide...</a><p>Secure Boot is still supported in that configuration, but with PK/db/dbx being part of the firmware configuration and updating them requiring a UEFI capsule update.</p>
]]></description><pubDate>Sun, 15 Mar 2026 16:29:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=47388973</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47388973</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47388973</guid></item><item><title><![CDATA[New comment by my123 in "Fedora 44 on the Raspberry Pi 5"]]></title><description><![CDATA[
<p>Note that UEFI doesn't mean supporting most of those.<p>UEFI without runtime UEFI variable writes is a thing, and that configuration is incompatible with mokutil.</p>
]]></description><pubDate>Sun, 15 Mar 2026 11:23:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=47386328</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47386328</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47386328</guid></item><item><title><![CDATA[New comment by my123 in "Fedora 44 on the Raspberry Pi 5"]]></title><description><![CDATA[
<p>U-Boot nowadays speaks UEFI :) (and so does LK)<p>New Android devices all use a UEFI bootloader: <a href="https://source.android.com/docs/core/architecture/bootloader/generic-bootloader/gbl-dev" rel="nofollow">https://source.android.com/docs/core/architecture/bootloader...</a></p>
]]></description><pubDate>Sat, 14 Mar 2026 23:23:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=47382405</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47382405</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47382405</guid></item><item><title><![CDATA[New comment by my123 in "Wired headphone sales are exploding"]]></title><description><![CDATA[
<p>Side note: USB 3 Lightning did exist on iPad Pros.</p>
]]></description><pubDate>Sat, 14 Mar 2026 20:06:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=47380618</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47380618</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47380618</guid></item><item><title><![CDATA[New comment by my123 in "Prefix sums at gigabytes per second with ARM NEON"]]></title><description><![CDATA[
<p>SME2 is restricted in scope to matrix multiply workloads and isn't really designed for anything else.<p>The point of streaming SVE is to have a way to pre/post process data on the way in or out of a matrix multiply.<p>A list that I have around of chips which support various levels of SVE:<p>For SVE(1) deployment, chips that have it:
- Fujitsu A64fx
- AWS Graviton3<p>SVE2:
- Snapdragon X2, 8/8 Elite Gen 5 and later
- MediaTek Dimensity 9000 and later
- NVIDIA Tegra Thor and later, NVIDIA "N1" or later (GB10 is an "N1x" SKU)
- Samsung Exynos 2200 or later
- AWS Graviton4, Microsoft Cobalt 100, Google Axion (and newer chips)
- CIX P1<p>SME(1) instead of SME2:<p>- Snapdragon X2, 8/8 Elite Gen 5<p>SME2:<p>- Apple M4, A18 and later
- Samsung Exynos 2600 
- MediaTek Dimensity 9500<p>Note that the Snapdragon 8/8 Elite Gen 5 and X2 support sve2 but not svebitperm.</p>
]]></description><pubDate>Fri, 13 Mar 2026 12:40:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=47363705</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47363705</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47363705</guid></item><item><title><![CDATA[New comment by my123 in "Prefix sums at gigabytes per second with ARM NEON"]]></title><description><![CDATA[
<p>Qualcomm was odd like that for a long time yeah.<p>And yes the Gen 5 chips (8, 8 Elite and X2) do implement SVE2 and SME.</p>
]]></description><pubDate>Fri, 13 Mar 2026 12:32:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=47363607</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47363607</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47363607</guid></item><item><title><![CDATA[New comment by my123 in "A new Oracle Solaris Common Build Environment (CBE) release"]]></title><description><![CDATA[
<p>AmpereOne M with MTE is out nowadays</p>
]]></description><pubDate>Tue, 10 Mar 2026 19:12:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=47327594</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47327594</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47327594</guid></item><item><title><![CDATA[New comment by my123 in "MacBook Air with M5"]]></title><description><![CDATA[
<p>LPCAMM2 is more present on business/high end machines unfortunately. It's not an Intel restriction.</p>
]]></description><pubDate>Tue, 03 Mar 2026 22:26:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=47239975</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47239975</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47239975</guid></item><item><title><![CDATA[New comment by my123 in "Arm's Cortex X925: Reaching Desktop Performance"]]></title><description><![CDATA[
<p>Google outright has worst in class SoCs on both CPU and GPU unfortunately.<p>If you want something more perf competitive, pick Dimensity, Exynos, or Snapdragon.</p>
]]></description><pubDate>Tue, 03 Mar 2026 22:24:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=47239953</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47239953</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47239953</guid></item><item><title><![CDATA[New comment by my123 in "Arm's Cortex X925: Reaching Desktop Performance"]]></title><description><![CDATA[
<p>> While the parent article shows AMD Zen 5 having significantly better results in floating-point SPEC CPU2017, these benchmark results are still misleading, because in properly optimized for AVX-512 applications the difference between Zen 5 and Cortex-X925 would be much greater. I have no idea how SPEC has been compiled by the author of the article, but the floating-point results are not consistent with programs optimized for Zen 5.<p>The arithmetic intensity of most SPECfp subtests is quite low. You see this wall because it ends up reaching bandwidth limitations long before running out of compute on cores with beefy SIMD.</p>
]]></description><pubDate>Tue, 03 Mar 2026 12:07:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=47231230</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47231230</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47231230</guid></item><item><title><![CDATA[New comment by my123 in "AVX2 is slower than SSE2-4.x under Windows ARM emulation"]]></title><description><![CDATA[
<p>It took until Alder Lake-N for the atom-grade stuff to have AVX2 across the board.</p>
]]></description><pubDate>Wed, 18 Feb 2026 15:55:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=47062371</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47062371</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47062371</guid></item><item><title><![CDATA[New comment by my123 in "AVX2 is slower than SSE2-4.x under Windows ARM emulation"]]></title><description><![CDATA[
<p>The Oryon 3rd gen in the Snapdragon X2 has SVE2 (as does NVIDIA N1x, currently pre-launched of sorts on the DGX Spark)</p>
]]></description><pubDate>Wed, 18 Feb 2026 15:54:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=47062347</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=47062347</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47062347</guid></item><item><title><![CDATA[Latest VirtualBox Code Begins Supporting KVM Back End]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.phoronix.com/news/VirtualBox-Upstream-With-KVM">https://www.phoronix.com/news/VirtualBox-Upstream-With-KVM</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46912777">https://news.ycombinator.com/item?id=46912777</a></p>
<p>Points: 7</p>
<p># Comments: 0</p>
]]></description><pubDate>Fri, 06 Feb 2026 13:47:28 +0000</pubDate><link>https://www.phoronix.com/news/VirtualBox-Upstream-With-KVM</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=46912777</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46912777</guid></item><item><title><![CDATA[New comment by my123 in "Demystifying ARM SME to Optimize General Matrix Multiplications"]]></title><description><![CDATA[
<p>Section VII.3 has:<p>> Libraries such as BLIS [19] lack SME
support and are therefore excluded from comparison.</p>
]]></description><pubDate>Sat, 31 Jan 2026 21:00:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=46840791</link><dc:creator>my123</dc:creator><comments>https://news.ycombinator.com/item?id=46840791</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46840791</guid></item></channel></rss>