dpibreak.git/src/platform/linux/rxring.rs, branch masterSystem-wide DPI circumvention with minimal configuration
http://git.dilluti0n.com/dpibreak.git/atom?h=master2026-07-10T08:32:39Zlinux: rxring: fix lifetime issue for Pkt::net()2026-07-10T08:32:39Zdilluti0nhskim@dilluti0n.com2026-07-10T08:32:39Zurn:sha1:47971046d4cc8e4964a207758831f6820a805859
Result should be unusable after Pkt is dropped, not rx is dropped.
linux: rxring: fix tp_snaplen treated as the L3 length instead L22026-07-10T07:59:23Zdilluti0nhskim@dilluti0n.com2026-07-10T07:59:23Zurn:sha1:062ae9fe82f9cc2256cbdbfdc5f00e387314f49a
While solving it, introduced Pkt abstraction so that advance() is
automatically executed upon dropping it.
This change still allows access to mmapped pointers within the
Pkt.net() and enables the addition of other slice fields (such as mac)
later.
linux: rxring: guard div0 as EINVAL2026-07-09T07:15:31Zdilluti0nhskim@dilluti0n.com2026-07-09T07:15:31Zurn:sha1:fb0f1e7ad8e7e191f66ffcdd712afcda9181dfa9linux: rxring: fix u32 overflow on ring_size init2026-07-09T07:07:30Zdilluti0nhskim@dilluti0n.com2026-07-09T07:07:08Zurn:sha1:758399909f58de6548f66f1ae6fe25c905fddc9flinux: rxring: fix possable memory ordering issue2026-07-09T06:27:32Zdilluti0nhskim@dilluti0n.com2026-07-09T06:26:46Zurn:sha1:96478e1940e7055cd9fcdb26824eae799a8ae62blinux: rxring: make tp_* series configurable2026-07-06T15:12:39Zdilluti0nhskim@dilluti0n.com2026-07-06T15:01:07Zurn:sha1:8a69237243c045800996846b6c2db4dd5f7383f6linux: implement socket and mmap wrapper for rxring2026-07-06T15:12:23Zdilluti0nhskim@dilluti0n.com2026-07-06T14:34:20Zurn:sha1:2a024e886547bdcddd9633d77baa0d25c5d3bb07
Here mmap/munmap wrapper remain unsafe since mmap returns a pointer
causes a memory leak when munmap is not called while dropping, and
munmap has strict rule (PAGE_SIZE aligned) for addr defined on
munmap(2).
linux: add PACKET_RX_RING to libc_s::setsockopt apstraction2026-07-06T13:05:28Zdilluti0nhskim@dilluti0n.com2026-07-06T13:05:28Zurn:sha1:e9db8ad3caf4c28f301ef7c6be72f9617cc490b5
The existing implementation was unsafe because UB could occur if a
user-space pointer referenced by the struct sock_fprog was incorrectly
passed. Rust safe model allows pointer creation and makes
dereferencing unsafe. In this case, dereferencing happenes in
kernel-space, Rust cannot guarantee this. So it must be handled
separately.
linux: add safe abstraction for setsockopt(SO_ATTACH_FILTER)2026-07-06T11:11:59Zdilluti0nhskim@dilluti0n.com2026-07-06T11:07:33Zurn:sha1:ca1737e17103d767691786b5bbf6de491dcc75a3
Treating optval as just a &[u8] in setsockopt() is not appropriate for
usage patterns where a struct is put into optval. Rust treats casting
a struct to &[u8] as unsafe.
linux: rxring: nit: add SPDX header2026-03-15T04:27:57Zdilluti0nhskimse1@gmail.com2026-03-15T04:27:57Zurn:sha1:57ff0e7c3279935f3879691d794bf5c8f798e5c0