What is Hackers' Pub?

Hackers' Pub is a place for software engineers to share their knowledge and experience with each other. It's also an ActivityPub-enabled social network, so you can follow your favorite hackers in the fediverse and get their latest posts in your feed.

0
0
0
0
0
0
0
0
0
amos @fasterthanlime@hachyderm.io
Public

I just spent the entire morning debugging a MacOS sandboxing issue that turns out to have been "I said no to a permission prompt two days ago and this actually broke the entire VFS and I can't find where to reset it." Wowee mac dev.

0
0
0
0
0
0
1
0
0
1
0
0
dch :flantifa: :flan_hacker: @dch@bsd.network
Public
Shared by Stefano Marinelli

anybody know how to tell what the *actual* checksum chosen by OpenZFS is, on FreeBSD?

When a zpool is created by bsdinstall, it defaults to checksum=on, and a micro benchmark is done to choose the appropriate checksum (from fletcher4, sha256, sha512, skein, blake3).

I would like to know what that chosen checksum is.

openzfs.github.io/openzfs-docs provides instructions for ZoL but its not matched on FreeBSD.
Sysctl shows only:

$ sysctl -a |egrep -i 'blake3|fletcher|sha2|sha5'
vfs.zfs.fletcher_4_impl: [fastest] scalar superscalar superscalar4 sse2 ssse3 avx2 avx512f avx512bw
vfs.zfs.sha512_impl: cycle [fastest] generic x64 avx avx2
vfs.zfs.sha256_impl: cycle [fastest] generic x64 ssse3 avx avx2 shani
vfs.zfs.blake3_impl: cycle [fastest] generic sse2 sse41 avx2 avx512

Which is not very helpful!

Inspecting zdb shows that the überblock uses fletcher4, if so that's a bad choice for almost all the machines I have, vs one of the more modern CPU-accelerated ones.

Uberblock:
magic = 0000000000bab10c
version = 5000
txg = 29818863
guid_sum = 12865431551488305392
timestamp = 1770722160 UTC = Tue Feb 10 11:16:00 2026
bp = DVA[0]=<0:1ef37817000:1000> DVA[1]=<0:1f26e5c1000:1000> DVA[2]=<0:8c341bf000:1000> [L0 DMU objset] fletcher4 uncompressed unencrypted LE contiguous unique triple size=1000L/1000P birth=29818863L/29818863P fill=75881 cksum=00000003035e0207:00000baf31d8f848:0016adf241101a15:1d6126f678dd7a91

Checksums and Their Use in ZFS — OpenZFS documentation

openzfs.github.io

0
0
0
0
0
0
0
0
0
0
0
0
Fesshole 🧻 @fesshole@mastodon.social
Public

We don't send you surveys to find out how satisfied you were with the driver who put your parcel in the bin just before it was emptied. We just need to maximise the opportunities for you to accept cookies so we can track you and sell data.

0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
kazuhito @kazuhito@vivaldi.net
Public

Favorite Photo: Patagonia Scale by Miroslav Matusinsky flic.kr/p/2rVKfdV

Patagonia Scale

A single human figure, wrapped in red, facing a glacial lake and a wall of stone that refuses to explain itself. The water is cold, opaque, ancient. The peaks above disappear into cloud, as if even they had something to hide. Nothing here is dramatic on purpose. The scale does all the talking. This is the moment when perspective resets — not because the world changes, but because you finally shut up long enough to see it.

www.flickr.com · Flickr

0
0
0
dch :flantifa: :flan_hacker: @dch@bsd.network
Public

anybody know how to tell what the *actual* checksum chosen by OpenZFS is, on FreeBSD?

When a zpool is created by bsdinstall, it defaults to checksum=on, and a micro benchmark is done to choose the appropriate checksum (from fletcher4, sha256, sha512, skein, blake3).

I would like to know what that chosen checksum is.

openzfs.github.io/openzfs-docs provides instructions for ZoL but its not matched on FreeBSD.
Sysctl shows only:

$ sysctl -a |egrep -i 'blake3|fletcher|sha2|sha5'
vfs.zfs.fletcher_4_impl: [fastest] scalar superscalar superscalar4 sse2 ssse3 avx2 avx512f avx512bw
vfs.zfs.sha512_impl: cycle [fastest] generic x64 avx avx2
vfs.zfs.sha256_impl: cycle [fastest] generic x64 ssse3 avx avx2 shani
vfs.zfs.blake3_impl: cycle [fastest] generic sse2 sse41 avx2 avx512

Which is not very helpful!

Inspecting zdb shows that the überblock uses fletcher4, if so that's a bad choice for almost all the machines I have, vs one of the more modern CPU-accelerated ones.

Uberblock:
magic = 0000000000bab10c
version = 5000
txg = 29818863
guid_sum = 12865431551488305392
timestamp = 1770722160 UTC = Tue Feb 10 11:16:00 2026
bp = DVA[0]=<0:1ef37817000:1000> DVA[1]=<0:1f26e5c1000:1000> DVA[2]=<0:8c341bf000:1000> [L0 DMU objset] fletcher4 uncompressed unencrypted LE contiguous unique triple size=1000L/1000P birth=29818863L/29818863P fill=75881 cksum=00000003035e0207:00000baf31d8f848:0016adf241101a15:1d6126f678dd7a91

Checksums and Their Use in ZFS — OpenZFS documentation

openzfs.github.io

0
0
0
tesaguri 🦀🦝 @tesaguri@fedibird.com
Public

[Bug] Redirects are not followed if request URL is not RFC-3986 compliant · Issue #2875 · seanmonstar/reqwest
github.com/seanmonstar/reqwest

野生のHTTPサーバのガバなあー

Example #[test] fn test() { tokio::runtime::Builder::new_current_thread() .enable_all() .build() .unwrap() .block_on(async { let client = reqwest::ClientBuilder::new() .http1_only() .https_only(tru...

[Bug] Redirects are not followed if request URL is not RFC-3986 compliant · Issue #2875 · seanmonstar/reqwest

Example #[test] fn test() { tokio::runtime::Builder::new_current_thread() .enable_all() .build() .unwrap() .block_on(async { let client = reqwest::ClientBuilder::new() .http1_only() .https_only(tru...

github.com · GitHub

0
0
0
dnamag @dnamag@daydream.ink
Public
Shared by tooearly

DNAMAG은 새로운 것들을 디깅하고
아카이빙하는 대중문화 매거진입니다.

DNAMAG은
인스타, 트위터, 블루스카이,
스레드, 마스토돈, 미스키 등의
플랫폼에서 연재하며

유니스쿼드를 통해
독자들과 아카이빙 합니다.

(타래에 이어서)

DNAMAG은 새로운 것들을 디깅하고 아카이빙하는 대중문화 매거진입니다.
0
0
1
1
1
0
0
0
0
3
0
0
0
0
0
0
0
0
robb @dlx@mastodon.social
Public

How do I show a confirmation dialog from "within” a menu like so but not in a toolbar?

Menu {
Button {
…
}
}
.confirmationDialog {
…
}

presents the dialog as a popover pointing at the close menu and

Menu {
Button {
…
}
.confirmationDialog {
…
}
}

doesn't work at all?

A partial screenshot of a new message being composed in Mail on iPhone, with the action sheet open after choosing to cancel the message. The action sheet presents choices to delete the draft or save the draft.
0
0
0
0
0
0
1
0
0
ploum @ploum@mamot.fr
Public

Twitter, Discord illustrent que, oui, vous allez à un moment être forcé de partir.

Alors écoutez ceux qui vous préviennent depuis des années parce que mieux vaut partir selon votre propre rhythme.

Au plus tard vous partez, au plus vous allez accepter des conditions de pire en pire. Et au plus vous retardez la migration des autres.

Car si vous restez sur MACHIN parce que "tout le monde y est", tout le monde y est justement parce que vous y êtes en premier lieu.

0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
1
0
1
0
0
0
1
0
0
1
0
0
dnamag @dnamag@daydream.ink
Public

DNAMAG은 새로운 것들을 디깅하고
아카이빙하는 대중문화 매거진입니다.

DNAMAG은
인스타, 트위터, 블루스카이,
스레드, 마스토돈, 미스키 등의
플랫폼에서 연재하며

유니스쿼드를 통해
독자들과 아카이빙 합니다.

(타래에 이어서)

DNAMAG은 새로운 것들을 디깅하고 아카이빙하는 대중문화 매거진입니다.
0
0
1
1
0
0
0
0
0
2
0
0