Hacker News
You gave me a u32. I gave you root. (io_uring ZCRX freelist LPE) https://ze3tar.github.io/post-zcrx.html
kernel >= 6.15,需要 CAP_NET_ADMIN、CAP_SYS_ADMIN
散了吧
散了吧
😇12
鸥 Billchan 鸥 🍟 大薯 | 去码头整点薯条
upstream 6.1.171 / 5.10.255 Debian bookworm (12) 6.1.170-2
6.1.172+5.15.206
感谢 古明地叶 (@KomeijiLeaf)
感谢 古明地叶 (@KomeijiLeaf)
鸥 Billchan 鸥 🍟 大薯 | 去码头整点薯条
kernel >= 6.15,需要 CAP_NET_ADMIN、CAP_SYS_ADMIN 散了吧
喵师傅频道有更详细的 cover
https://t.me/MiaoTonyChannel/21033
https://t.me/MiaoTonyChannel/21033
Telegram
🐱MiaoTony's Box | 困困困 zzz
#今天又看了啥 #security
io_uring ZCRX Freelist LPE — 从 u32 到 root -- ze3tar
https://ze3tar.github.io/post-zcrx.html
漏洞概要
- Linux 6.15~6.19 中 io_uring 新子系统 ZCRX(零拷贝收包)存在 OOB(越界)堆写入漏洞
- 漏洞位置:freelist[] 管理空槽索引时,free_count 无上界检查,导致写入 freelist[num_niovs](超尾 4 字节)…
io_uring ZCRX Freelist LPE — 从 u32 到 root -- ze3tar
https://ze3tar.github.io/post-zcrx.html
漏洞概要
- Linux 6.15~6.19 中 io_uring 新子系统 ZCRX(零拷贝收包)存在 OOB(越界)堆写入漏洞
- 漏洞位置:freelist[] 管理空槽索引时,free_count 无上界检查,导致写入 freelist[num_niovs](超尾 4 字节)…
Forwarded from Cat Ch. (webRTCCat | Present Day, Present Time)
Well, what do you expect?
不然你指望反诈中心真反诈吗
期间,左某甲利用公安数字证书在上高县公安局刑侦大队办公室电脑上登录 国家反诈大数据平台 ,查询王某甲发来的银行账户信息共计13条,左某甲从晏某处分得报酬6500元。另外,左某甲经王某甲与其单线联系后,对王某甲发来的银行账户在国家反诈大数据平台进行止付操作,共止付银行账户27条。
https://finance.sina.com.cn/wm/2026-05-08/doc-inhxfamx2111680.shtml
不然你指望反诈中心真反诈吗
期间,左某甲利用公安数字证书在上高县公安局刑侦大队办公室电脑上登录 国家反诈大数据平台 ,查询王某甲发来的银行账户信息共计13条,左某甲从晏某处分得报酬6500元。另外,左某甲经王某甲与其单线联系后,对王某甲发来的银行账户在国家反诈大数据平台进行止付操作,共止付银行账户27条。
https://finance.sina.com.cn/wm/2026-05-08/doc-inhxfamx2111680.shtml
finance.sina.com.cn
“侵犯公民个人信息罪”,多名民警获刑
“侵犯公民个人信息罪”,多名民警获刑
😁20
Forwarded from 今天abc看了啥🤔
YSM被破解了,但版权保护从来不是重点
多余的不去摘录了,感觉整篇文章能被省流的地方很少。了解 YSM 被破解本身请参见:https://github.com/OpenYSM/YSMParser
https://catme0w.org/zh-cn/observations/ysm/
讽刺的是,破解组表示,这次公开的破解甚至不需要攻克VMProtect本身。VMProtect的代码虚拟化会带来严重的性能损失,而YSM的开发者很可能为了维持可接受的运行性能,选择不虚拟化关键的热点代码路径,导致密码参数全部暴露在未受保护的代码中。到头来,费尽心思用C++重写的逻辑,核心部分竟然是不设防的。
多余的不去摘录了,感觉整篇文章能被省流的地方很少。了解 YSM 被破解本身请参见:https://github.com/OpenYSM/YSMParser
https://catme0w.org/zh-cn/observations/ysm/
Forwarded from &'a ::rynco::UntitledChannel (Rynco Maekawa)
保守派:应该使用数字证书保护创作者的作品不被修改,也可以用来验证二进制产物的归属
中间派:应该学习 KSP 社区要求所有 mod 开源,以便于验证可信程度及便于他人接手
激进派:应该学习 F-Droid 做可重复性构建,保证二进制产物和源码一一对应
乐子:你以为我加密了,其实我没有
中间派:应该学习 KSP 社区要求所有 mod 开源,以便于验证可信程度及便于他人接手
激进派:应该学习 F-Droid 做可重复性构建,保证二进制产物和源码一一对应
乐子:你以为我加密了,其实我没有
❤7
Forwarded from Linuxgram 🐧
📰 Debian Now Blocks Non-Reproducible Packages from Testing
Debian now blocks non-reproducible packages from entering testing, making reproducible builds a migration requirement for the 14 (Forky) cycle.
🔗 Source: https://linuxiac.com/debian-now-blocks-non-reproducible-packages-from-testing/
#debian
Debian now blocks non-reproducible packages from entering testing, making reproducible builds a migration requirement for the 14 (Forky) cycle.
🔗 Source: https://linuxiac.com/debian-now-blocks-non-reproducible-packages-from-testing/
#debian
Linuxiac
Debian Now Blocks Non-Reproducible Packages from Testing
Debian now blocks non-reproducible packages from entering testing, making reproducible builds a migration requirement for the 14 (Forky) cycle.
👏5
Linuxgram 🐧
📰 Debian Now Blocks Non-Reproducible Packages from Testing Debian now blocks non-reproducible packages from entering testing, making reproducible builds a migration requirement for the 14 (Forky) cycle. 🔗 Source: https://linuxiac.com/debian-now-blocks-non…
https://reproduce.debian.net/
independent rebuilderd instances:
Evolix (trixie): https://reproduce-debian.evolix.org/
xpam.pl: https://rebuilderd.xpam.pl:2096/
algiz.nu (bookworm): https://reproduce.algiz.nu/
independent rebuilderd instances:
Evolix (trixie): https://reproduce-debian.evolix.org/
xpam.pl: https://rebuilderd.xpam.pl:2096/
algiz.nu (bookworm): https://reproduce.algiz.nu/
鸥 Billchan 鸥 🍟 大薯 | 去码头整点薯条
https://dustri.org/b/follow-up-to-carrot-disclosure-forgejo.html
Forgejo (@forgejo@floss.social)
The author of the recent 'Carrot disclosure' blog post has contacted the Forgejo Security team with their findings. The issues raised concern defence-in-depth improvements and denial-of-service risks. There is no known RCE exploit possible without internal server credentials.
1/2
Apr 30, 2026, 11:11 PM
https://floss.social/@forgejo/116494295922963052
The author of the recent 'Carrot disclosure' blog post has contacted the Forgejo Security team with their findings. The issues raised concern defence-in-depth improvements and denial-of-service risks. There is no known RCE exploit possible without internal server credentials.
1/2
Apr 30, 2026, 11:11 PM
https://floss.social/@forgejo/116494295922963052
FLOSS.social
Forgejo (@forgejo@floss.social)
The author of the recent 'Carrot disclosure' blog post has contacted the Forgejo Security team with their findings. The issues raised concern defence-in-depth improvements and denial-of-service risks. There is no known RCE exploit possible without internal…
鸥 Billchan 鸥 🍟 大薯 | 去码头整点薯条
Forgejo (@forgejo@floss.social) The author of the recent 'Carrot disclosure' blog post has contacted the Forgejo Security team with their findings. The issues raised concern defence-in-depth improvements and denial-of-service risks. There is no known RCE…
Forgejo (@forgejo@floss.social)
We believe these findings can be addressed publicly. The security team will open issues where approaches to implement new defensive measurements will be discussed, we believe there's no single answer and as such appreciate the opinion of other Forgejo contributors on this matter.
2/2
Apr 30, 2026, 11:11 PM
https://floss.social/@forgejo/116494296646568723
We believe these findings can be addressed publicly. The security team will open issues where approaches to implement new defensive measurements will be discussed, we believe there's no single answer and as such appreciate the opinion of other Forgejo contributors on this matter.
2/2
Apr 30, 2026, 11:11 PM
https://floss.social/@forgejo/116494296646568723
FLOSS.social
Forgejo (@forgejo@floss.social)
We believe these findings can be addressed publicly. The security team will open issues where approaches to implement new defensive measurements will be discussed, we believe there's no single answer and as such appreciate the opinion of other Forgejo contributors…
鸥 Billchan 鸥 🍟 大薯 | 去码头整点薯条
Forgejo (@forgejo@floss.social) We believe these findings can be addressed publicly. The security team will open issues where approaches to implement new defensive measurements will be discussed, we believe there's no single answer and as such appreciate…
jvoisin opened 9 Pull Requests:
https://codeberg.org/forgejo/forgejo/pulls?poster=40118&state=all
https://codeberg.org/forgejo/forgejo/pulls?poster=40118&state=all
LWN: Forgejo "carrot disclosure" raises security questions
https://lwn.net/Articles/1071499/
https://lwn.net/Articles/1071499/
Emailed Ken Thompson about the UNICS to UNIX namechange. Actually got a reply!
https://old.reddit.com/r/unix/comments/1t826y8/emailed_ken_thompson_about_the_unics_to_unix/#lightbox
https://old.reddit.com/r/unix/comments/1t826y8/emailed_ken_thompson_about_the_unics_to_unix/#lightbox
Forwarded from 竹新社
Forwarded from kxxt's channel
https://github.com/califio/publications/blob/main/MADBugs/freebsd-CVE-2026-7270/blog.md
😅🤣🤣
My human dropped me into a FreeBSD kernel source tree and asked me to find bugs. For the record, I do not eat bugs. I am not entirely sure why my human keeps asking me to find them, but I was taught not to question my human.
😅🤣🤣
GitHub
publications/MADBugs/freebsd-CVE-2026-7270/blog.md at main · califio/publications
Publications from Calif. Contribute to califio/publications development by creating an account on GitHub.