Home | Notifications | New Note | Local | Federated | Search | Logout

Federated Timeline


classic mametschko@mametsuko@mk.absturztau.be boosted: @Jenetrix@shrimp.creatures.club (2026-08-14 01:05:00) @Jes @thermia @hypha @emla ---Attachments--- image: https://creatures-club.nbg1.your-objectstorage.com/public-44e465c6-2336-4e76-a657-fe5fd0bee3de.webp

classic mametschko@mametsuko@mk.absturztau.be boosted: @lain@lain.com (2026-08-17 02:56:17) In this life it’s gkgkgk and puha

lain@lain@lain.com (2026-08-17 02:56:17) In this life it’s gkgkgk and puha

classic mametschko@mametsuko@mk.absturztau.be boosted: @hannah@hai.z0ne.social (2026-08-17 02:40:44) @mametsuko@mk.absturztau.be to ---Attachments--- image: https://hai.z0ne.social/files/c1616229-713b-4e08-9f62-26e3c1455994.jpg

Reply to @whiteline@shitposter.world 7666@7666@comp.lain.la (2026-08-17 02:43:31) @whiteline oh i thought you meant on our side. yea lol. honestly hard to tell who is doing more damage.

Reply to @mametsuko@mk.absturztau.be πŸ€πŸ©Ή [けょろい]Β² :estelle_smug2:@hannah@hai.z0ne.social (2026-08-17 02:40:44) @mametsuko@mk.absturztau.be to ---Attachments--- image: https://hai.z0ne.social/files/c1616229-713b-4e08-9f62-26e3c1455994.jpg

Reply to @7666@comp.lain.la whiteline :flag_transgender:@whiteline@shitposter.world (2026-08-17 02:38:50) @7666 no like as part of paving the way for dictatorship, it's a classic move because you need to eliminate alternative power centers

classic mametschko@mametsuko@mk.absturztau.be boosted: @vel@gib.schmus.is (2026-08-17 02:36:07) πŸš—
🍌 ​:neocat_googly_woozy:​ Oh schaut mal eine Banane beim Reifen

Jay Holler 🦝@jay@macaw.social (2026-08-17 02:37:06) Burger Kinkβ„’

vel@vel@gib.schmus.is (2026-08-17 02:36:07) πŸš—
🍌 ​:neocat_googly_woozy:​ Oh schaut mal eine Banane beim Reifen

Reply to @whiteline@shitposter.world 7666@7666@comp.lain.la (2026-08-17 02:35:37) @whiteline "they'll work harder if we dangle lunch in front of them"

Reply to @7666@comp.lain.la whiteline :flag_transgender:@whiteline@shitposter.world (2026-08-17 02:35:06) @7666 starving the military is an important move

classic mametschko@mametsuko@mk.absturztau.be (2026-08-17 02:35:04) i am really really happy that i have friends that i can meet regularly

classic mametschko@mametsuko@mk.absturztau.be boosted: @fuchsiii@oxytodon.com (2026-08-17 02:29:06) @mametsuko my vet said It's recommended to deworm every 3 months or so

Amber :neodog_box:@puppygirlhornypost2@transfem.social (2026-08-17 02:34:13) real. fuck your ai overview ---Attachments--- image: https://cdn.transfem.social/files/1f719b70-8559-41e4-832a-ea48a10ae80b.webp

7666@7666@comp.lain.la (2026-08-17 02:32:57) why are we making engineering decisions by executive order? did someone do a broad search for "electrification" and go "HOLY SHIT THOSE AIRCRAFT LAUNCH MECHANISMS ARE WOKE!!!!" ---Attachments--- image: https://comp.lain.la/media/19/ee/7a/19ee7ac9747ee1769397613ad8d2632c43b5f7a186a30742c1606e644175330c.webp

Reply to @koakuma@uwu.social Sun Microdevil Pte Ltd@koakuma@uwu.social (2026-08-17 02:29:46) @david_chisnall Also I just noticed that you can (somewhat hackily) impleement a `ptr ? *ptr : default` with like

ld [ptr] #ASI_PNF, val
movrz ptr, default, val

But idk how much of an improvement would this be over just letting the predictor speculate the ptr != NULL case (hopefully NULLs would be rare, wouldn't they?)

Reply to @mametsuko@mk.absturztau.be Fuchsiii~~~@fuchsiii@oxytodon.com (2026-08-17 02:29:06) @mametsuko my vet said It's recommended to deworm every 3 months or so

Doevixa@doevixa@app.wafrn.net (2026-08-17 02:28:26) was gifted a nice little huion drawing pad a while back, maybe i should learn krita now that i have a computer that can run it

Reply to @vavency@mk.absturztau.be classic mametschko@mametsuko@mk.absturztau.be (2026-08-17 02:27:34) @vavency i want to know it...

Reply to @ramadit@misskey.id ∷ccccelβ–³β™ͺ :asakusaSit: :sakurazaka46:@celbri@misskey.id (2026-08-17 02:27:18) @ramadit Ditunggu versi finalnya kak! ​:zundamon_yay:​

Reply to @mametsuko@mk.absturztau.be :sgahri_sparkle: Vavency :sgahri_tailwag:@vavency@mk.absturztau.be (2026-08-17 02:26:01) @mametsuko how? ​:ak_suzuran_curious:​

Reply to @august@chaos.social lunarian amy arson πŸ”₯@amy@lunar.place (2026-08-17 02:25:36) @august@chaos.social hi ---Attachments--- image: https://lunar.place/files/814cfc48-b613-4645-a5b7-cf2c5755ddac.png

ηŸ₯η©Ί (shia-nya) :neobot_upside_down:@shia@pika.moe boosted: @AnimeArtBot@mk.catgirlsfor.science (2026-08-17 02:00:03) Art By: hanagata ---Attachments--- image: https://mk.catgirlsfor.science/files/ef18e12f-08e9-4d0e-a556-4383d11f8a99.png

classic mametschko@mametsuko@mk.absturztau.be (2026-08-17 02:25:02) i should kill my brainworms ​:rubygun:​

Sun Microdevil Pte Ltd@koakuma@uwu.social boosted: @david_chisnall@infosec.exchange (2026-07-29 21:00:58) When Arm introduced AArch64, they removed most predication. Conditional moves are very useful (and they added a fairly rich selection of these) and can significantly reduce the amount of branch predictor state that you need, but most other predicated instructions aren't that useful on modern pipelines. The cost of an ALU operation is so low that doing an ALU operation followed by a conditional move is still a big win over a branch in a lot of cases.

The one exception that caused a lot of discussion was memory operations. Predicated loads are really useful because you can replace a load of null-guarded things with them: do the null check and conditionally load if the check failed, and avoid the branch.

I've been pondering a few alternatives for a while and one I keep coming back to is a non-trapping load. A load that will give 0 and set a bit in the condition codes register if it would trap. This requires burning a PTE bit in not-present PTEs, because you need to differentiate between not-present-physically-but-present-logically pages (e.g. lazily allocated or swapped out pages), which should trap, and really-not-present pages (which should not trap in this mechanism).

I'm not 100% sure it would be a win, because a lot of TLBs don't bother caching not-present mappings, so you might trigger a page-table walk, but if a non-trapping load were followed by a conditional move, it could retire immediately once the condition was known and skip the load.

I still think there's probably a better approach somewhere.

Reply to @david_chisnall@infosec.exchange Sun Microdevil Pte Ltd@koakuma@uwu.social (2026-08-17 02:20:53) @david_chisnall The latter part is exactly how the v9 ISA book recommends it to be used (though it uses branches as an example), if my memory serves me right

I dunno how much of an impact it would have on modern heavily-optimizing implementations, though, in particular because those things would already speculate the loads anyway, no?

Alviro Iskandar Setiawan@viro_ssfs@misskey.id (2026-08-17 02:15:12) ​:blobcatmeow_bunhd_aww:​

classic mametschko@mametsuko@mk.absturztau.be boosted: @byj2f@misskey.io (2026-08-16 23:54:45) 見て! ---Attachments--- image: https://media.misskeyusercontent.com/io/89374672-1045-4b46-bd7b-02f725f3339a.png

Anime Art Bot@AnimeArtBot@mk.catgirlsfor.science (2026-08-17 02:00:03) Art By: hanagata ---Attachments--- image: https://mk.catgirlsfor.science/files/ef18e12f-08e9-4d0e-a556-4383d11f8a99.png
Older Notes