sphinx.addnodesdocument)}( rawsourcechildren]( translations LanguagesNode)}(hhh](h pending_xref)}(hhh]docutils.nodesTextChinese (Simplified)}parenthsba attributes}(ids]classes]names]dupnames]backrefs] refdomainstdreftypedoc reftarget*/translations/zh_CN/admin-guide/sysctl/netmodnameN classnameN refexplicitutagnamehhh ubh)}(hhh]hChinese (Traditional)}hh2sbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/zh_TW/admin-guide/sysctl/netmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hItalian}hhFsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/it_IT/admin-guide/sysctl/netmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hJapanese}hhZsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/ja_JP/admin-guide/sysctl/netmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hKorean}hhnsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/ko_KR/admin-guide/sysctl/netmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hPortuguese (Brazilian)}hhsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/pt_BR/admin-guide/sysctl/netmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hSpanish}hhsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/sp_SP/admin-guide/sysctl/netmodnameN classnameN refexplicituh1hhh ubeh}(h]h ]h"]h$]h&]current_languageEnglishuh1h hh _documenthsourceNlineNubhsection)}(hhh](htitle)}(h Documentation for /proc/sys/net/h]h Documentation for /proc/sys/net/}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhhhD/var/lib/git/docbuild/linux/Documentation/admin-guide/sysctl/net.rsthKubh paragraph)}(h Copyrighth]h Copyright}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(hCopyright (c) 1999h]hCopyright (c) 1999}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh block_quote)}(hH- Terrehon Bowden - Bodo Bauer h]h bullet_list)}(hhh](h list_item)}(h&Terrehon Bowden h]h)}(hhh](hTerrehon Bowden <}(hhhhhNhNubh reference)}(hterrehon@pacbell.neth]hterrehon@pacbell.net}(hjhhhNhNubah}(h]h ]h"]h$]h&]refurimailto:terrehon@pacbell.netuh1jhhubh>}(hhhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhK hhubah}(h]h ]h"]h$]h&]uh1hhhubh)}(hBodo Bauer h]h)}(hBodo Bauer h](h Bodo Bauer <}(hj'hhhNhNubj)}(hbb@ricochet.neth]hbb@ricochet.net}(hj/hhhNhNubah}(h]h ]h"]h$]h&]refurimailto:bb@ricochet.netuh1jhj'ubh>}(hj'hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhK hj#ubah}(h]h ]h"]h$]h&]uh1hhhubeh}(h]h ]h"]h$]h&]bullet-uh1hhhhK hhubah}(h]h ]h"]h$]h&]uh1hhhhK hhhhubh)}(hCopyright (c) 2000h]hCopyright (c) 2000}(hj]hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK hhhhubh)}(h)- Jorge Nerin h]h)}(hhh]h)}(h'Jorge Nerin h]h)}(h&Jorge Nerin h](h Jorge Nerin <}(hjvhhhNhNubj)}(hcomandante@zaralinux.comh]hcomandante@zaralinux.com}(hj~hhhNhNubah}(h]h ]h"]h$]h&]refurimailto:comandante@zaralinux.comuh1jhjvubh>}(hjvhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhjrubah}(h]h ]h"]h$]h&]uh1hhjoubah}(h]h ]h"]h$]h&]jUjVuh1hhhhKhjkubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(hCopyright (c) 2009h]hCopyright (c) 2009}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(h"- Shen Feng h]h)}(hhh]h)}(h Shen Feng h]h)}(hShen Feng h](h Shen Feng <}(hjhhhNhNubj)}(hshen@cn.fujitsu.comh]hshen@cn.fujitsu.com}(hjhhhNhNubah}(h]h ]h"]h$]h&]refurimailto:shen@cn.fujitsu.comuh1jhjubh>}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]jUjVuh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(h;For general info and legal blurb, please look in index.rst.h]h;For general info and legal blurb, please look in index.rst.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh transition)}(hN------------------------------------------------------------------------------h]h}(h]h ]h"]h$]h&]uh1jhhhKhhhhubh)}(hJThis file contains the documentation for the sysctl files in /proc/sys/neth]hJThis file contains the documentation for the sysctl files in /proc/sys/net}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(hThe interface to the networking parts of the kernel is located in /proc/sys/net. The following table shows all possible subdirectories. You may see only some of them, depending on your kernel's configuration.h]hThe interface to the networking parts of the kernel is located in /proc/sys/net. The following table shows all possible subdirectories. You may see only some of them, depending on your kernel’s configuration.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(h'Table : Subdirectories in /proc/sys/neth]h'Table : Subdirectories in /proc/sys/net}(hj-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK hhhhubh)}(hX[========= =================== = ========== =================== Directory Content Directory Content ========= =================== = ========== =================== 802 E802 protocol mptcp Multipath TCP appletalk Appletalk protocol netfilter Network Filter bridge Bridging tipc TIPC core General parameter unix Unix domain sockets ethernet Ethernet protocol vsock VSOCK sockets ipv4 IP version 4 x25 X.25 protocol ipv6 IP version 6 ========= =================== = ========== =================== h]htable)}(hhh]htgroup)}(hhh](hcolspec)}(hhh]h}(h]h ]h"]h$]h&]colwidthK uh1jIhjFubjJ)}(hhh]h}(h]h ]h"]h$]h&]colwidthKuh1jIhjFubjJ)}(hhh]h}(h]h ]h"]h$]h&]colwidthKuh1jIhjFubjJ)}(hhh]h}(h]h ]h"]h$]h&]colwidthK uh1jIhjFubjJ)}(hhh]h}(h]h ]h"]h$]h&]colwidthKuh1jIhjFubhthead)}(hhh]hrow)}(hhh](hentry)}(hhh]h)}(h Directoryh]h Directory}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK#hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(hContenth]hContent}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK#hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(h Directoryh]h Directory}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK#hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(hContenth]hContent}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK#hjubah}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1j}hjFubhtbody)}(hhh](j)}(hhh](j)}(hhh]h)}(h802h]h802}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK%hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(h E802 protocolh]h E802 protocol}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK%hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(hmptcph]hmptcp}(hj<hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK%hj9ubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(h Multipath TCPh]h Multipath TCP}(hjShhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK%hjPubah}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh](j)}(hhh]h)}(h appletalkh]h appletalk}(hjshhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK&hjpubah}(h]h ]h"]h$]h&]uh1jhjmubj)}(hhh]h)}(hAppletalk protocolh]hAppletalk protocol}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK&hjubah}(h]h ]h"]h$]h&]uh1jhjmubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjmubj)}(hhh]h)}(h netfilterh]h netfilter}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK&hjubah}(h]h ]h"]h$]h&]uh1jhjmubj)}(hhh]h)}(hNetwork Filterh]hNetwork Filter}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK&hjubah}(h]h ]h"]h$]h&]uh1jhjmubeh}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh](j)}(hhh]h)}(hbridgeh]hbridge}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK'hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(hBridgingh]hBridging}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK'hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(htipch]htipc}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK'hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(hTIPCh]hTIPC}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK'hj,ubah}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh](j)}(hhh]h)}(hcoreh]hcore}(hjOhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK(hjLubah}(h]h ]h"]h$]h&]uh1jhjIubj)}(hhh]h)}(hGeneral parameterh]hGeneral parameter}(hjfhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK(hjcubah}(h]h ]h"]h$]h&]uh1jhjIubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjIubj)}(hhh]h)}(hunixh]hunix}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK(hjubah}(h]h ]h"]h$]h&]uh1jhjIubj)}(hhh]h)}(hUnix domain socketsh]hUnix domain sockets}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK(hjubah}(h]h ]h"]h$]h&]uh1jhjIubeh}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh](j)}(hhh]h)}(hetherneth]hethernet}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK)hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(hEthernet protocolh]hEthernet protocol}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK)hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(hvsockh]hvsock}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK)hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(h VSOCK socketsh]h VSOCK sockets}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK)hjubah}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh](j)}(hhh]h)}(hipv4h]hipv4}(hj+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK*hj(ubah}(h]h ]h"]h$]h&]uh1jhj%ubj)}(hhh]h)}(h IP version 4h]h IP version 4}(hjBhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK*hj?ubah}(h]h ]h"]h$]h&]uh1jhj%ubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhj%ubj)}(hhh]h)}(hx25h]hx25}(hjbhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK*hj_ubah}(h]h ]h"]h$]h&]uh1jhj%ubj)}(hhh]h)}(h X.25 protocolh]h X.25 protocol}(hjyhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK*hjvubah}(h]h ]h"]h$]h&]uh1jhj%ubeh}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh](j)}(hhh]h)}(hipv6h]hipv6}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK+hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h)}(h IP version 6h]h IP version 6}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK+hjubah}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjubj)}(hhh]h}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjFubeh}(h]h ]h"]h$]h&]colsKuh1jDhjAubah}(h]h ]h"]h$]h&]uh1j?hj;ubah}(h]h ]h"]h$]h&]uh1hhhhK"hhhhubh)}(hhh](h)}(h,1. /proc/sys/net/core - Network core optionsh]h,1. /proc/sys/net/core - Network core options}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhK/ubh)}(hhh](h)}(hbpf_jit_enableh]hbpf_jit_enable}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhK2ubh)}(hXlThis enables the BPF Just in Time (JIT) compiler. BPF is a flexible and efficient infrastructure allowing to execute bytecode at various hook points. It is used in a number of Linux kernel subsystems such as networking (e.g. XDP, tc), tracing (e.g. kprobes, uprobes, tracepoints) and security (e.g. seccomp). LLVM has a BPF back end that can compile restricted C into a sequence of BPF instructions. After program load through bpf(2) and passing a verifier in the kernel, a JIT will then translate these BPF proglets into native CPU instructions. There are two flavors of JITs, the newer eBPF JIT currently supported on:h]hXlThis enables the BPF Just in Time (JIT) compiler. BPF is a flexible and efficient infrastructure allowing to execute bytecode at various hook points. It is used in a number of Linux kernel subsystems such as networking (e.g. XDP, tc), tracing (e.g. kprobes, uprobes, tracepoints) and security (e.g. seccomp). LLVM has a BPF back end that can compile restricted C into a sequence of BPF instructions. After program load through bpf(2) and passing a verifier in the kernel, a JIT will then translate these BPF proglets into native CPU instructions. There are two flavors of JITs, the newer eBPF JIT currently supported on:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK4hjhhubh)}(hu- x86_64 - x86_32 - arm64 - arm32 - ppc64 - ppc32 - sparc64 - mips64 - s390x - riscv64 - riscv32 - loongarch64 - arc h]h)}(hhh](h)}(hx86_64h]h)}(hj7h]hx86_64}(hj9hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK>hj5ubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(hx86_32h]h)}(hjNh]hx86_32}(hjPhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK?hjLubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(harm64h]h)}(hjeh]harm64}(hjghhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK@hjcubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(harm32h]h)}(hj|h]harm32}(hj~hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKAhjzubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(hppc64h]h)}(hjh]hppc64}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKBhjubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(hppc32h]h)}(hjh]hppc32}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKChjubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(hsparc64h]h)}(hjh]hsparc64}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKDhjubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(hmips64h]h)}(hjh]hmips64}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKEhjubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(hs390xh]h)}(hjh]hs390x}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKFhjubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(hriscv64h]h)}(hjh]hriscv64}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKGhjubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(hriscv32h]h)}(hjh]hriscv32}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKHhjubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(h loongarch64h]h)}(hj4h]h loongarch64}(hj6hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKIhj2ubah}(h]h ]h"]h$]h&]uh1hhj2ubh)}(harc h]h)}(harch]harc}(hjMhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKJhjIubah}(h]h ]h"]h$]h&]uh1hhj2ubeh}(h]h ]h"]h$]h&]jUjVuh1hhhhK>hj.ubah}(h]h ]h"]h$]h&]uh1hhhhK>hjhhubh)}(h8And the older cBPF JIT supported on the following archs:h]h8And the older cBPF JIT supported on the following archs:}(hjmhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKLhjhhubh)}(h- mips - sparc h]h)}(hhh](h)}(hmipsh]h)}(hjh]hmips}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKNhjubah}(h]h ]h"]h$]h&]uh1hhjubh)}(hsparc h]h)}(hsparch]hsparc}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKOhjubah}(h]h ]h"]h$]h&]uh1hhjubeh}(h]h ]h"]h$]h&]jUjVuh1hhhhKNhj{ubah}(h]h ]h"]h$]h&]uh1hhhhKNhjhhubh)}(hXeBPF JITs are a superset of cBPF JITs, meaning the kernel will migrate cBPF instructions into eBPF instructions and then JIT compile them transparently. Older cBPF JITs can only translate tcpdump filters, seccomp rules, etc, but not mentioned eBPF programs loaded through bpf(2).h]hXeBPF JITs are a superset of cBPF JITs, meaning the kernel will migrate cBPF instructions into eBPF instructions and then JIT compile them transparently. Older cBPF JITs can only translate tcpdump filters, seccomp rules, etc, but not mentioned eBPF programs loaded through bpf(2).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKQhjhhubh)}(hValues:h]hValues:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKWhjhhubh)}(h- 0 - disable the JIT (default value) - 1 - enable the JIT - 2 - enable the JIT and ask the compiler to emit traces on kernel log. h]h)}(hhh](h)}(h#0 - disable the JIT (default value)h]h)}(hjh]h#0 - disable the JIT (default value)}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKYhjubah}(h]h ]h"]h$]h&]uh1hhjubh)}(h1 - enable the JITh]h)}(hjh]h1 - enable the JIT}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKZhjubah}(h]h ]h"]h$]h&]uh1hhjubh)}(hF2 - enable the JIT and ask the compiler to emit traces on kernel log. h]h)}(hE2 - enable the JIT and ask the compiler to emit traces on kernel log.h]hE2 - enable the JIT and ask the compiler to emit traces on kernel log.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK[hjubah}(h]h ]h"]h$]h&]uh1hhjubeh}(h]h ]h"]h$]h&]jUjVuh1hhhhKYhjubah}(h]h ]h"]h$]h&]uh1hhhhKYhjhhubeh}(h]bpf-jit-enableah ]h"]bpf_jit_enableah$]h&]uh1hhjhhhhhK2ubh)}(hhh](h)}(hbpf_jit_hardenh]hbpf_jit_harden}(hj=hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj:hhhhhK^ubh)}(hThis enables hardening for the BPF JIT compiler. Supported are eBPF JIT backends. Enabling hardening trades off performance, but can mitigate JIT spraying.h]hThis enables hardening for the BPF JIT compiler. Supported are eBPF JIT backends. Enabling hardening trades off performance, but can mitigate JIT spraying.}(hjKhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK`hj:hhubh)}(hValues:h]hValues:}(hjYhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKdhj:hhubh)}(h- 0 - disable JIT hardening (default value) - 1 - enable JIT hardening for unprivileged users only - 2 - enable JIT hardening for all users h]h)}(hhh](h)}(h)0 - disable JIT hardening (default value)h]h)}(hjph]h)0 - disable JIT hardening (default value)}(hjrhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKfhjnubah}(h]h ]h"]h$]h&]uh1hhjkubh)}(h41 - enable JIT hardening for unprivileged users onlyh]h)}(hjh]h41 - enable JIT hardening for unprivileged users only}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKghjubah}(h]h ]h"]h$]h&]uh1hhjkubh)}(h'2 - enable JIT hardening for all users h]h)}(h&2 - enable JIT hardening for all usersh]h&2 - enable JIT hardening for all users}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhjubah}(h]h ]h"]h$]h&]uh1hhjkubeh}(h]h ]h"]h$]h&]jUjVuh1hhhhKfhjgubah}(h]h ]h"]h$]h&]uh1hhhhKfhj:hhubh)}(htwhere "privileged user" in this context means a process having CAP_BPF or CAP_SYS_ADMIN in the root user name space.h]hxwhere “privileged user” in this context means a process having CAP_BPF or CAP_SYS_ADMIN in the root user name space.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKjhj:hhubeh}(h]bpf-jit-hardenah ]h"]bpf_jit_hardenah$]h&]uh1hhjhhhhhK^ubh)}(hhh](h)}(hbpf_jit_kallsymsh]hbpf_jit_kallsyms}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhKnubh)}(hX"When BPF JIT compiler is enabled, then compiled images are unknown addresses to the kernel, meaning they neither show up in traces nor in /proc/kallsyms. This enables export of these addresses, which can be used for debugging/tracing. If bpf_jit_harden is enabled, this feature is disabled.h]hX"When BPF JIT compiler is enabled, then compiled images are unknown addresses to the kernel, meaning they neither show up in traces nor in /proc/kallsyms. This enables export of these addresses, which can be used for debugging/tracing. If bpf_jit_harden is enabled, this feature is disabled.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKphjhhubh)}(hValues :h]hValues :}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKvhjhhubh)}(hm- 0 - disable JIT kallsyms export (default value) - 1 - enable JIT kallsyms export for privileged users only h]h)}(hhh](h)}(h/0 - disable JIT kallsyms export (default value)h]h)}(hj h]h/0 - disable JIT kallsyms export (default value)}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKxhj ubah}(h]h ]h"]h$]h&]uh1hhj ubh)}(h91 - enable JIT kallsyms export for privileged users only h]h)}(h81 - enable JIT kallsyms export for privileged users onlyh]h81 - enable JIT kallsyms export for privileged users only}(hj% hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKyhj! ubah}(h]h ]h"]h$]h&]uh1hhj ubeh}(h]h ]h"]h$]h&]jUjVuh1hhhhKxhj ubah}(h]h ]h"]h$]h&]uh1hhhhKxhjhhubeh}(h]bpf-jit-kallsymsah ]h"]bpf_jit_kallsymsah$]h&]uh1hhjhhhhhKnubh)}(hhh](h)}(h bpf_jit_limith]h bpf_jit_limit}(hjP hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjM hhhhhK|ubh)}(hThis enforces a global limit for memory allocations to the BPF JIT compiler in order to reject unprivileged JIT requests once it has been surpassed. bpf_jit_limit contains the value of the global limit in bytes.h]hThis enforces a global limit for memory allocations to the BPF JIT compiler in order to reject unprivileged JIT requests once it has been surpassed. bpf_jit_limit contains the value of the global limit in bytes.}(hj^ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK~hjM hhubeh}(h] bpf-jit-limitah ]h"] bpf_jit_limitah$]h&]uh1hhjhhhhhK|ubh)}(hhh](h)}(h dev_weighth]h dev_weight}(hjw hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjt hhhhhKubh)}(hThe maximum number of packets that kernel can handle on a NAPI interrupt, it's a Per-CPU variable. For drivers that support LRO or GRO_HW, a hardware aggregated packet is counted as one packet in this context.h]hThe maximum number of packets that kernel can handle on a NAPI interrupt, it’s a Per-CPU variable. For drivers that support LRO or GRO_HW, a hardware aggregated packet is counted as one packet in this context.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjt hhubh)}(h Default: 64h]h Default: 64}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjt hhubeh}(h] dev-weightah ]h"] dev_weightah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(hdev_weight_rx_biash]hdev_weight_rx_bias}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(hX9RPS (e.g. RFS, aRFS) processing is competing with the registered NAPI poll function of the driver for the per softirq cycle netdev_budget. This parameter influences the proportion of the configured netdev_budget that is spent on RPS based packet processing during RX softirq cycles. It is further meant for making current dev_weight adaptable for asymmetric CPU needs on RX/TX side of the network stack. (see dev_weight_tx_bias) It is effective on a per CPU basis. Determination is based on dev_weight and is calculated multiplicative (dev_weight * dev_weight_rx_bias).h]hX9RPS (e.g. RFS, aRFS) processing is competing with the registered NAPI poll function of the driver for the per softirq cycle netdev_budget. This parameter influences the proportion of the configured netdev_budget that is spent on RPS based packet processing during RX softirq cycles. It is further meant for making current dev_weight adaptable for asymmetric CPU needs on RX/TX side of the network stack. (see dev_weight_tx_bias) It is effective on a per CPU basis. Determination is based on dev_weight and is calculated multiplicative (dev_weight * dev_weight_rx_bias).}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubh)}(h Default: 1h]h Default: 1}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubeh}(h]dev-weight-rx-biasah ]h"]dev_weight_rx_biasah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(hdev_weight_tx_biash]hdev_weight_tx_bias}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(hScales the maximum number of packets that can be processed during a TX softirq cycle. Effective on a per CPU basis. Allows scaling of current dev_weight for asymmetric net stack processing needs. Be careful to avoid making TX softirq processing a CPU hog.h]hScales the maximum number of packets that can be processed during a TX softirq cycle. Effective on a per CPU basis. Allows scaling of current dev_weight for asymmetric net stack processing needs. Be careful to avoid making TX softirq processing a CPU hog.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubh)}(hECalculation is based on dev_weight (dev_weight * dev_weight_tx_bias).h]hECalculation is based on dev_weight (dev_weight * dev_weight_tx_bias).}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubh)}(h Default: 1h]h Default: 1}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubeh}(h]dev-weight-tx-biasah ]h"]dev_weight_tx_biasah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(h default_qdisch]h default_qdisc}(hj$ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj! hhhhhKubh)}(hXThe default queuing discipline to use for network devices. This allows overriding the default of pfifo_fast with an alternative. Since the default queuing discipline is created without additional parameters so is best suited to queuing disciplines that work well without configuration like stochastic fair queue (sfq), CoDel (codel) or fair queue CoDel (fq_codel). Don't use queuing disciplines like Hierarchical Token Bucket or Deficit Round Robin which require setting up classes and bandwidths. Note that physical multiqueue interfaces still use mq as root qdisc, which in turn uses this default for its leaves. Virtual devices (like e.g. lo or veth) ignore this setting and instead default to noqueue.h]hXThe default queuing discipline to use for network devices. This allows overriding the default of pfifo_fast with an alternative. Since the default queuing discipline is created without additional parameters so is best suited to queuing disciplines that work well without configuration like stochastic fair queue (sfq), CoDel (codel) or fair queue CoDel (fq_codel). Don’t use queuing disciplines like Hierarchical Token Bucket or Deficit Round Robin which require setting up classes and bandwidths. Note that physical multiqueue interfaces still use mq as root qdisc, which in turn uses this default for its leaves. Virtual devices (like e.g. lo or veth) ignore this setting and instead default to noqueue.}(hj2 hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj! hhubh)}(hDefault: pfifo_fasth]hDefault: pfifo_fast}(hj@ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj! hhubeh}(h] default-qdiscah ]h"] default_qdiscah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(h busy_readh]h busy_read}(hjY hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjV hhhhhKubh)}(hXLow latency busy poll timeout for socket reads. (needs CONFIG_NET_RX_BUSY_POLL) Approximate time in us to busy loop waiting for packets on the device queue. This sets the default value of the SO_BUSY_POLL socket option. Can be set or overridden per socket by setting socket option SO_BUSY_POLL, which is the preferred method of enabling. If you need to enable the feature globally via sysctl, a value of 50 is recommended.h]hXLow latency busy poll timeout for socket reads. (needs CONFIG_NET_RX_BUSY_POLL) Approximate time in us to busy loop waiting for packets on the device queue. This sets the default value of the SO_BUSY_POLL socket option. Can be set or overridden per socket by setting socket option SO_BUSY_POLL, which is the preferred method of enabling. If you need to enable the feature globally via sysctl, a value of 50 is recommended.}(hjg hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjV hhubh)}(hWill increase power usage.h]hWill increase power usage.}(hju hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjV hhubh)}(hDefault: 0 (off)h]hDefault: 0 (off)}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjV hhubeh}(h] busy-readah ]h"] busy_readah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(h busy_pollh]h busy_poll}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(hXLow latency busy poll timeout for poll and select. (needs CONFIG_NET_RX_BUSY_POLL) Approximate time in us to busy loop waiting for events. Recommended value depends on the number of sockets you poll on. For several sockets 50, for several hundreds 100. For more than that you probably want to use epoll. Note that only sockets with SO_BUSY_POLL set will be busy polled, so you want to either selectively set SO_BUSY_POLL on those sockets or set sysctl.net.busy_read globally.h]hXLow latency busy poll timeout for poll and select. (needs CONFIG_NET_RX_BUSY_POLL) Approximate time in us to busy loop waiting for events. Recommended value depends on the number of sockets you poll on. For several sockets 50, for several hundreds 100. For more than that you probably want to use epoll. Note that only sockets with SO_BUSY_POLL set will be busy polled, so you want to either selectively set SO_BUSY_POLL on those sockets or set sysctl.net.busy_read globally.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubh)}(hWill increase power usage.h]hWill increase power usage.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubh)}(hDefault: 0 (off)h]hDefault: 0 (off)}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubeh}(h] busy-pollah ]h"] busy_pollah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(h mem_pcpu_rsvh]h mem_pcpu_rsv}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(hMPer-cpu reserved forward alloc cache size in page units. Default 1MB per CPU.h]hMPer-cpu reserved forward alloc cache size in page units. Default 1MB per CPU.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubeh}(h] mem-pcpu-rsvah ]h"] mem_pcpu_rsvah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(hbypass_prot_memh]hbypass_prot_mem}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(hSkip charging socket buffers to the global per-protocol memory accounting controlled by net.ipv4.tcp_mem, net.ipv4.udp_mem, etc.h]hSkip charging socket buffers to the global per-protocol memory accounting controlled by net.ipv4.tcp_mem, net.ipv4.udp_mem, etc.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubh)}(hDefault: 0 (off)h]hDefault: 0 (off)}(hj" hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubeh}(h]bypass-prot-memah ]h"]bypass_prot_memah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(h rmem_defaulth]h rmem_default}(hj; hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj8 hhhhhKubh)}(h:The default setting of the socket receive buffer in bytes.h]h:The default setting of the socket receive buffer in bytes.}(hjI hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj8 hhubeh}(h] rmem-defaultah ]h"] rmem_defaultah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(hrmem_maxh]hrmem_max}(hjb hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj_ hhhhhKubh)}(h0The maximum receive socket buffer size in bytes.h]h0The maximum receive socket buffer size in bytes.}(hjp hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj_ hhubh)}(hDefault: 4194304h]hDefault: 4194304}(hj~ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj_ hhubeh}(h]rmem-maxah ]h"]rmem_maxah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(hrps_default_maskh]hrps_default_mask}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(hlThe default RPS CPU mask used on newly created network devices. An empty mask means RPS disabled by default.h]hlThe default RPS CPU mask used on newly created network devices. An empty mask means RPS disabled by default.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubeh}(h]rps-default-maskah ]h"]rps_default_maskah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(htstamp_allow_datah]htstamp_allow_data}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(hAllow processes to receive tx timestamps looped together with the original packet contents. If disabled, transmit timestamp requests from unprivileged processes are dropped unless socket option SOF_TIMESTAMPING_OPT_TSONLY is set.h]hAllow processes to receive tx timestamps looped together with the original packet contents. If disabled, transmit timestamp requests from unprivileged processes are dropped unless socket option SOF_TIMESTAMPING_OPT_TSONLY is set.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubh)}(hDefault: 1 (on)h]hDefault: 1 (on)}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubeh}(h]tstamp-allow-dataah ]h"]tstamp_allow_dataah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(h wmem_defaulth]h wmem_default}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(h9The default setting (in bytes) of the socket send buffer.h]h9The default setting (in bytes) of the socket send buffer.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj hhubeh}(h] wmem-defaultah ]h"] wmem_defaultah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(hwmem_maxh]hwmem_max}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubh)}(h-The maximum send socket buffer size in bytes.h]h-The maximum send socket buffer size in bytes.}(hj( hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj hhubh)}(hDefault: 4194304h]hDefault: 4194304}(hj6 hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj hhubeh}(h]wmem-maxah ]h"]wmem_maxah$]h&]uh1hhjhhhhhKubh)}(hhh](h)}(hmessage_burst and message_costh]hmessage_burst and message_cost}(hjO hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjL hhhhhMubh)}(hXThese parameters are used to limit the warning messages written to the kernel log from the networking code. They enforce a rate limit to make a denial-of-service attack impossible. A higher message_cost factor, results in fewer messages that will be written. Message_burst controls when messages will be dropped. The default settings limit warning messages to one every five seconds.h]hXThese parameters are used to limit the warning messages written to the kernel log from the networking code. They enforce a rate limit to make a denial-of-service attack impossible. A higher message_cost factor, results in fewer messages that will be written. Message_burst controls when messages will be dropped. The default settings limit warning messages to one every five seconds.}(hj] hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjL hhubeh}(h]message-burst-and-message-costah ]h"]message_burst and message_costah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(hwarningsh]hwarnings}(hjv hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjs hhhhhMubh)}(hThis sysctl is now unused.h]hThis sysctl is now unused.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjs hhubh)}(hThis was used to control console messages from the networking stack that occur because of problems on the network like duplicate address or bad checksums.h]hThis was used to control console messages from the networking stack that occur because of problems on the network like duplicate address or bad checksums.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjs hhubh)}(hwThese messages are now emitted at KERN_DEBUG and can generally be enabled and controlled by the dynamic_debug facility.h]hwThese messages are now emitted at KERN_DEBUG and can generally be enabled and controlled by the dynamic_debug facility.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjs hhubeh}(h]warningsah ]h"]warningsah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(h netdev_budgeth]h netdev_budget}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhMubh)}(hX.Maximum number of packets taken from all interfaces in one polling cycle (NAPI poll). In one polling cycle interfaces which are registered to polling are probed in a round-robin manner. Also, a polling cycle may not exceed netdev_budget_usecs microseconds, even if netdev_budget has not been exhausted.h]hX.Maximum number of packets taken from all interfaces in one polling cycle (NAPI poll). In one polling cycle interfaces which are registered to polling are probed in a round-robin manner. Also, a polling cycle may not exceed netdev_budget_usecs microseconds, even if netdev_budget has not been exhausted.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj hhubeh}(h] netdev-budgetah ]h"] netdev_budgetah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(hnetdev_budget_usecsh]hnetdev_budget_usecs}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhM%ubh)}(hMaximum number of microseconds in one NAPI polling cycle. Polling will exit when either netdev_budget_usecs have elapsed during the poll cycle or the number of packets processed reaches netdev_budget.h]hMaximum number of microseconds in one NAPI polling cycle. Polling will exit when either netdev_budget_usecs have elapsed during the poll cycle or the number of packets processed reaches netdev_budget.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM'hj hhubeh}(h]netdev-budget-usecsah ]h"]netdev_budget_usecsah$]h&]uh1hhjhhhhhM%ubh)}(hhh](h)}(hnetdev_max_backlogh]hnetdev_max_backlog}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhM,ubh)}(h}Maximum number of packets, queued on the INPUT side, when the interface receives packets faster than kernel can process them.h]h}Maximum number of packets, queued on the INPUT side, when the interface receives packets faster than kernel can process them.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM.hj hhubeh}(h]netdev-max-backlogah ]h"]netdev_max_backlogah$]h&]uh1hhjhhhhhM,ubh)}(hhh](h)}(hqdisc_max_bursth]hqdisc_max_burst}(hj. hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj+ hhhhhM2ubh)}(hOMaximum number of packets that can be temporarily stored before reaching qdisc.h]hOMaximum number of packets that can be temporarily stored before reaching qdisc.}(hj< hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM4hj+ hhubh)}(h Default: 1000h]h Default: 1000}(hjJ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM7hj+ hhubeh}(h]qdisc-max-burstah ]h"]qdisc_max_burstah$]h&]uh1hhjhhhhhM2ubh)}(hhh](h)}(hnetdev_rss_keyh]hnetdev_rss_key}(hjc hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj` hhhhhM:ubh)}(hRSS (Receive Side Scaling) enabled drivers use a host key that is randomly generated. Some user space might need to gather its content even if drivers do not provide ethtool -x support yet.h]hRSS (Receive Side Scaling) enabled drivers use a host key that is randomly generated. Some user space might need to gather its content even if drivers do not provide ethtool -x support yet.}(hjq hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM<hj` hhubh literal_block)}(hmyhost:~# cat /proc/sys/net/core/netdev_rss_key 84:50:f4:00:a8:15:d1:a7:e9:7f:1d:60:35:c7:47:25:42:97:74:ca:56:bb:b6:a1:d8: ... (256 bytes total)h]hmyhost:~# cat /proc/sys/net/core/netdev_rss_key 84:50:f4:00:a8:15:d1:a7:e9:7f:1d:60:35:c7:47:25:42:97:74:ca:56:bb:b6:a1:d8: ... (256 bytes total)}hj sbah}(h]h ]h"]h$]h&] xml:spacepreserveuh1j hhhMChj` hhubh)}(hTFile contains all nul bytes if no driver ever called netdev_rss_key_fill() function.h]hTFile contains all nul bytes if no driver ever called netdev_rss_key_fill() function.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMFhj` hhubhdefinition_list)}(hhh]hdefinition_list_item)}(hsNote: /proc/sys/net/core/netdev_rss_key contains 256 bytes of key, but many drivers only use 40 or 52 bytes of it. h](hterm)}(hNote:h]hNote:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j hhhMKhj ubh definition)}(hhh]h)}(hl/proc/sys/net/core/netdev_rss_key contains 256 bytes of key, but many drivers only use 40 or 52 bytes of it.h]hl/proc/sys/net/core/netdev_rss_key contains 256 bytes of key, but many drivers only use 40 or 52 bytes of it.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMJhj ubah}(h]h ]h"]h$]h&]uh1j hj ubeh}(h]h ]h"]h$]h&]uh1j hhhMKhj ubah}(h]h ]h"]h$]h&]uh1j hj` hhhhhNubj )}(hXmyhost:~# ethtool -x eth0 RX flow hash indirection table for eth0 with 8 RX ring(s): 0: 0 1 2 3 4 5 6 7 RSS hash key: 84:50:f4:00:a8:15:d1:a7:e9:7f:1d:60:35:c7:47:25:42:97:74:ca:56:bb:b6:a1:d8:43:e3:c9:0c:fd:17:55:c2:3a:4d:69:ed:f1:42:89h]hXmyhost:~# ethtool -x eth0 RX flow hash indirection table for eth0 with 8 RX ring(s): 0: 0 1 2 3 4 5 6 7 RSS hash key: 84:50:f4:00:a8:15:d1:a7:e9:7f:1d:60:35:c7:47:25:42:97:74:ca:56:bb:b6:a1:d8:43:e3:c9:0c:fd:17:55:c2:3a:4d:69:ed:f1:42:89}hj sbah}(h]h ]h"]h$]h&]j j uh1j hhhMOhj` hhubeh}(h]netdev-rss-keyah ]h"]netdev_rss_keyah$]h&]uh1hhjhhhhhM:ubh)}(hhh](h)}(hnetdev_tstamp_prequeueh]hnetdev_tstamp_prequeue}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhMVubh)}(hIf set to 0, RX packet timestamps can be sampled after RPS processing, when the target CPU processes packets. It might give some delay on timestamps, but permit to distribute the load on several cpus.h]hIf set to 0, RX packet timestamps can be sampled after RPS processing, when the target CPU processes packets. It might give some delay on timestamps, but permit to distribute the load on several cpus.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMXhj hhubh)}(hSIf set to 1 (default), timestamps are sampled as soon as possible, before queueing.h]hSIf set to 1 (default), timestamps are sampled as soon as possible, before queueing.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM\hj hhubeh}(h]netdev-tstamp-prequeueah ]h"]netdev_tstamp_prequeueah$]h&]uh1hhjhhhhhMVubh)}(hhh](h)}(hnetdev_unregister_timeout_secsh]hnetdev_unregister_timeout_secs}(hj-hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj*hhhhhM`ubh)}(hXUnregister network device timeout in seconds. This option controls the timeout (in seconds) used to issue a warning while waiting for a network device refcount to drop to 0 during device unregistration. A lower value may be useful during bisection to detect a leaked reference faster. A larger value may be useful to prevent false warnings on slow/loaded systems. Default value is 10, minimum 1, maximum 3600.h]hXUnregister network device timeout in seconds. This option controls the timeout (in seconds) used to issue a warning while waiting for a network device refcount to drop to 0 during device unregistration. A lower value may be useful during bisection to detect a leaked reference faster. A larger value may be useful to prevent false warnings on slow/loaded systems. Default value is 10, minimum 1, maximum 3600.}(hj;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMbhj*hhubeh}(h]netdev-unregister-timeout-secsah ]h"]netdev_unregister_timeout_secsah$]h&]uh1hhjhhhhhM`ubh)}(hhh](h)}(h skb_defer_maxh]h skb_defer_max}(hjThhhNhNubah}(h]h ]h"]h$]h&]uh1hhjQhhhhhMkubh)}(h[Max size (in skbs) of the per-cpu list of skbs being freed by the cpu which allocated them.h]h[Max size (in skbs) of the per-cpu list of skbs being freed by the cpu which allocated them.}(hjbhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMmhjQhhubh)}(h Default: 128h]h Default: 128}(hjphhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMphjQhhubeh}(h] skb-defer-maxah ]h"] skb_defer_maxah$]h&]uh1hhjhhhhhMkubh)}(hhh](h)}(h optmem_maxh]h optmem_max}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMsubh)}(hMaximum ancillary buffer size allowed per socket. Ancillary data is a sequence of struct cmsghdr structures with appended data. TCP tx zerocopy also uses optmem_max as a limit for its internal structures.h]hMaximum ancillary buffer size allowed per socket. Ancillary data is a sequence of struct cmsghdr structures with appended data. TCP tx zerocopy also uses optmem_max as a limit for its internal structures.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMuhjhhubh)}(hDefault : 128 KBh]hDefault : 128 KB}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMyhjhhubeh}(h] optmem-maxah ]h"] optmem_maxah$]h&]uh1hhjhhhhhMsubh)}(hhh](h)}(hfb_tunnels_only_for_init_neth]hfb_tunnels_only_for_init_net}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhM|ubh)}(hXControls if fallback tunnels (like tunl0, gre0, gretap0, erspan0, sit0, ip6tnl0, ip6gre0) are automatically created. There are 3 possibilities (a) value = 0; respective fallback tunnels are created when module is loaded in every net namespaces (backward compatible behavior). (b) value = 1; [kcmd value: initns] respective fallback tunnels are created only in init net namespace and every other net namespace will not have them. (c) value = 2; [kcmd value: none] fallback tunnels are not created when a module is loaded in any of the net namespace. Setting value to "2" is pointless after boot if these modules are built-in, so there is a kernel command-line option that can change this default. Please refer to Documentation/admin-guide/kernel-parameters.txt for additional details.h]hXControls if fallback tunnels (like tunl0, gre0, gretap0, erspan0, sit0, ip6tnl0, ip6gre0) are automatically created. There are 3 possibilities (a) value = 0; respective fallback tunnels are created when module is loaded in every net namespaces (backward compatible behavior). (b) value = 1; [kcmd value: initns] respective fallback tunnels are created only in init net namespace and every other net namespace will not have them. (c) value = 2; [kcmd value: none] fallback tunnels are not created when a module is loaded in any of the net namespace. Setting value to “2” is pointless after boot if these modules are built-in, so there is a kernel command-line option that can change this default. Please refer to Documentation/admin-guide/kernel-parameters.txt for additional details.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM~hjhhubh)}(hNot creating fallback tunnels gives control to userspace to create whatever is needed only and avoid creating devices which are redundant.h]hNot creating fallback tunnels gives control to userspace to create whatever is needed only and avoid creating devices which are redundant.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubh)}(h(Default : 0 (for compatibility reasons)h]h(Default : 0 (for compatibility reasons)}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h]fb-tunnels-only-for-init-netah ]h"]fb_tunnels_only_for_init_netah$]h&]uh1hhjhhhhhM|ubh)}(hhh](h)}(hdevconf_inherit_init_neth]hdevconf_inherit_init_net}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(hX Controls if a new network namespace should inherit all current settings under /proc/sys/net/{ipv4,ipv6}/conf/{all,default}/. By default, we keep the current behavior: for IPv4 we inherit all current settings from init_net and for IPv6 we reset all settings to default.h]hX Controls if a new network namespace should inherit all current settings under /proc/sys/net/{ipv4,ipv6}/conf/{all,default}/. By default, we keep the current behavior: for IPv4 we inherit all current settings from init_net and for IPv6 we reset all settings to default.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubh)}(hX8If set to 1, both IPv4 and IPv6 settings are forced to inherit from current ones in init_net. If set to 2, both IPv4 and IPv6 settings are forced to reset to their default values. If set to 3, both IPv4 and IPv6 settings are forced to inherit from current ones in the netns where this new netns has been created.h]hX8If set to 1, both IPv4 and IPv6 settings are forced to inherit from current ones in init_net. If set to 2, both IPv4 and IPv6 settings are forced to reset to their default values. If set to 3, both IPv4 and IPv6 settings are forced to inherit from current ones in the netns where this new netns has been created.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubh)}(h(Default : 0 (for compatibility reasons)h]h(Default : 0 (for compatibility reasons)}(hj+hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h]devconf-inherit-init-netah ]h"]devconf_inherit_init_netah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(htxrehashh]htxrehash}(hjDhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjAhhhhhMubh)}(hControls default hash rethink behaviour on socket when SO_TXREHASH option is set to SOCK_TXREHASH_DEFAULT (i. e. not overridden by setsockopt).h]hControls default hash rethink behaviour on socket when SO_TXREHASH option is set to SOCK_TXREHASH_DEFAULT (i. e. not overridden by setsockopt).}(hjRhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjAhhubh)}(hqIf set to 1 (default), hash rethink is performed on listening socket. If set to 0, hash rethink is not performed.h]hqIf set to 1 (default), hash rethink is performed on listening socket. If set to 0, hash rethink is not performed.}(hj`hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjAhhubeh}(h]txrehashah ]h"]txrehashah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(htxq_reselection_msh]htxq_reselection_ms}(hjyhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjvhhhhhMubh)}(hMControls how often (in ms) a busy connected flow can select another tx queue.h]hMControls how often (in ms) a busy connected flow can select another tx queue.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjvhhubh)}(hA resection is desirable when/if user thread has migrated and XPS would select a different queue. Same can occur without XPS if the flow hash has changed.h]hA resection is desirable when/if user thread has migrated and XPS would select a different queue. Same can occur without XPS if the flow hash has changed.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjvhhubh)}(hBut switching txq can introduce reorders, especially if the old queue is under high pressure. Modern TCP stacks deal well with reorders if they happen not too often.h]hBut switching txq can introduce reorders, especially if the old queue is under high pressure. Modern TCP stacks deal well with reorders if they happen not too often.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjvhhubh)}(h,To disable this feature, set the value to 0.h]h,To disable this feature, set the value to 0.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjvhhubh)}(hDefault : 1000h]hDefault : 1000}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjvhhubeh}(h]txq-reselection-msah ]h"]txq_reselection_msah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(hgro_normal_batchh]hgro_normal_batch}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(hX=Maximum number of the segments to batch up on output of GRO. When a packet exits GRO, either as a coalesced superframe or as an original packet which GRO has decided not to coalesce, it is placed on a per-NAPI list. This list is then passed to the stack when the number of segments reaches the gro_normal_batch limit.h]hX=Maximum number of the segments to batch up on output of GRO. When a packet exits GRO, either as a coalesced superframe or as an original packet which GRO has decided not to coalesce, it is placed on a per-NAPI list. This list is then passed to the stack when the number of segments reaches the gro_normal_batch limit.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h]gro-normal-batchah ]h"]gro_normal_batchah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(hhigh_order_alloc_disableh]hhigh_order_alloc_disable}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(hXBy default the allocator for page frags tries to use high order pages (order-3 on x86). While the default behavior gives good results in most cases, some users might have hit a contention in page allocations/freeing. This was especially true on older kernels (< 5.14) when high-order pages were not stored on per-cpu lists. This allows to opt-in for order-0 allocation instead but is now mostly of historical importance.h]hXBy default the allocator for page frags tries to use high order pages (order-3 on x86). While the default behavior gives good results in most cases, some users might have hit a contention in page allocations/freeing. This was especially true on older kernels (< 5.14) when high-order pages were not stored on per-cpu lists. This allows to opt-in for order-0 allocation instead but is now mostly of historical importance.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubh)}(h Default: 0h]h Default: 0}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h]high-order-alloc-disableah ]h"]high_order_alloc_disableah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(h:2. /proc/sys/net/unix - Parameters for Unix domain socketsh]h:2. /proc/sys/net/unix - Parameters for Unix domain sockets}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj1hhhhhMubh)}(hThere is only one file in this directory. unix_dgram_qlen limits the max number of datagrams queued in Unix domain socket's buffer. It will not take effect unless PF_UNIX flag is specified.h]hThere is only one file in this directory. unix_dgram_qlen limits the max number of datagrams queued in Unix domain socket’s buffer. It will not take effect unless PF_UNIX flag is specified.}(hjBhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj1hhubeh}(h]4proc-sys-net-unix-parameters-for-unix-domain-socketsah ]h"]:2. /proc/sys/net/unix - parameters for unix domain socketsah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(h%3. /proc/sys/net/ipv4 - IPV4 settingsh]h%3. /proc/sys/net/ipv4 - IPV4 settings}(hj[hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjXhhhhhMubh)}(hPlease see: Documentation/networking/ip-sysctl.rst and Documentation/admin-guide/sysctl/net.rst for descriptions of these entries.h]hPlease see: Documentation/networking/ip-sysctl.rst and Documentation/admin-guide/sysctl/net.rst for descriptions of these entries.}(hjihhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjXhhubeh}(h]proc-sys-net-ipv4-ipv4-settingsah ]h"]%3. /proc/sys/net/ipv4 - ipv4 settingsah$]h&]uh1hhjhhhhhMubh)}(hhh]h)}(h4. TIPCh]h4. TIPC}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubah}(h]tipcah ]h"]4. tipcah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(h tipc_rmemh]h tipc_rmem}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(hThe TIPC protocol now has a tunable for the receive memory, similar to the tcp_rmem - i.e. a vector of 3 INTEGERs: (min, default, max)h]hThe TIPC protocol now has a tunable for the receive memory, similar to the tcp_rmem - i.e. a vector of 3 INTEGERs: (min, default, max)}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubj )}(hE# cat /proc/sys/net/tipc/tipc_rmem 4252725 34021800 68043600 #h]hE# cat /proc/sys/net/tipc/tipc_rmem 4252725 34021800 68043600 #}hjsbah}(h]h ]h"]h$]h&]j j uh1j hhhMhjhhubh)}(hX!The max value is set to CONN_OVERLOAD_LIMIT, and the default and min values are scaled (shifted) versions of that same value. Note that the min value is not at this point in time used in any meaningful way, but the triplet is preserved in order to be consistent with things like tcp_rmem.h]hX!The max value is set to CONN_OVERLOAD_LIMIT, and the default and min values are scaled (shifted) versions of that same value. Note that the min value is not at this point in time used in any meaningful way, but the triplet is preserved in order to be consistent with things like tcp_rmem.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h] tipc-rmemah ]h"] tipc_rmemah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(h named_timeouth]h named_timeout}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(hXTIPC name table updates are distributed asynchronously in a cluster, without any form of transaction handling. This means that different race scenarios are possible. One such is that a name withdrawal sent out by one node and received by another node may arrive after a second, overlapping name publication already has been accepted from a third node, although the conflicting updates originally may have been issued in the correct sequential order. If named_timeout is nonzero, failed topology updates will be placed on a defer queue until another event arrives that clears the error, or until the timeout expires. Value is in milliseconds.h]hXTIPC name table updates are distributed asynchronously in a cluster, without any form of transaction handling. This means that different race scenarios are possible. One such is that a name withdrawal sent out by one node and received by another node may arrive after a second, overlapping name publication already has been accepted from a third node, although the conflicting updates originally may have been issued in the correct sequential order. If named_timeout is nonzero, failed topology updates will be placed on a defer queue until another event arrives that clears the error, or until the timeout expires. Value is in milliseconds.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h] named-timeoutah ]h"] named_timeoutah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(h&6. /proc/sys/net/vsock - VSOCK socketsh]h&6. /proc/sys/net/vsock - VSOCK sockets}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(hX.VSOCK sockets (AF_VSOCK) provide communication between virtual machines and their hosts. The behavior of VSOCK sockets in a network namespace is determined by the namespace's mode (``global`` or ``local``), which controls how CIDs (Context IDs) are allocated and how sockets interact across namespaces.h](hVSOCK sockets (AF_VSOCK) provide communication between virtual machines and their hosts. The behavior of VSOCK sockets in a network namespace is determined by the namespace’s mode (}(hjhhhNhNubhliteral)}(h ``global``h]hglobal}(hjhhhNhNuba@h}(h]h ]h"]h$]h&]uh1jhjubh or }(hjhhhNhNubj)}(h ``local``h]hlocal}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubhb), which controls how CIDs (Context IDs) are allocated and how sockets interact across namespaces.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjhhubeh}(h] proc-sys-net-vsock-vsock-socketsah ]h"]&6. /proc/sys/net/vsock - vsock socketsah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(hns_modeh]hns_mode}(hjRhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjOhhhhhMubh)}(hdRead-only. Reports the current namespace's mode, set at namespace creation and immutable thereafter.h]hfRead-only. Reports the current namespace’s mode, set at namespace creation and immutable thereafter.}(hj`hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjOhhubh)}(hValues:h]hValues:}(hjnhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM hjOhhubh)}(hXV- ``global`` - the namespace shares system-wide CID allocation and its sockets can reach any VM or socket in any global namespace. Sockets in this namespace cannot reach sockets in local namespaces. - ``local`` - the namespace has private CID allocation and its sockets can only connect to VMs or sockets within the same namespace. h]h)}(hhh](h)}(h``global`` - the namespace shares system-wide CID allocation and its sockets can reach any VM or socket in any global namespace. Sockets in this namespace cannot reach sockets in local namespaces.h]h)}(h``global`` - the namespace shares system-wide CID allocation and its sockets can reach any VM or socket in any global namespace. Sockets in this namespace cannot reach sockets in local namespaces.h](j)}(h ``global``h]hglobal}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh - the namespace shares system-wide CID allocation and its sockets can reach any VM or socket in any global namespace. Sockets in this namespace cannot reach sockets in local namespaces.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1hhjubh)}(h``local`` - the namespace has private CID allocation and its sockets can only connect to VMs or sockets within the same namespace. h]h)}(h``local`` - the namespace has private CID allocation and its sockets can only connect to VMs or sockets within the same namespace.h](j)}(h ``local``h]hlocal}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubhy - the namespace has private CID allocation and its sockets can only connect to VMs or sockets within the same namespace.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1hhjubeh}(h]h ]h"]h$]h&]jUjVuh1hhhhMhj|ubah}(h]h ]h"]h$]h&]uh1hhhhMhjOhhubh)}(h'The init_net mode is always ``global``.h](hThe init_net mode is always }(hjhhhNhNubj)}(h ``global``h]hglobal}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjOhhubeh}(h]ns-modeah ]h"]ns_modeah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(h child_ns_modeh]h child_ns_mode}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubh)}(hControls what mode newly created child namespaces will inherit. At namespace creation, ``ns_mode`` is inherited from the parent's ``child_ns_mode``. The initial value matches the namespace's own ``ns_mode``.h](hWControls what mode newly created child namespaces will inherit. At namespace creation, }(hjhhhNhNubj)}(h ``ns_mode``h]hns_mode}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh" is inherited from the parent’s }(hjhhhNhNubj)}(h``child_ns_mode``h]h child_ns_mode}(hj.hhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh2. The initial value matches the namespace’s own }(hjhhhNhNubj)}(h ``ns_mode``h]hns_mode}(hj@hhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjhhubh)}(hValues:h]hValues:}(hjXhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjhhubh)}(hX+- ``global`` - child namespaces will share system-wide CID allocation and their sockets will be able to reach any VM or socket in any global namespace. - ``local`` - child namespaces will have private CID allocation and their sockets will only be able to connect within their own namespace. h]h)}(hhh](h)}(h``global`` - child namespaces will share system-wide CID allocation and their sockets will be able to reach any VM or socket in any global namespace.h]h)}(h``global`` - child namespaces will share system-wide CID allocation and their sockets will be able to reach any VM or socket in any global namespace.h](j)}(h ``global``h]hglobal}(hjuhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjqubh - child namespaces will share system-wide CID allocation and their sockets will be able to reach any VM or socket in any global namespace.}(hjqhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM!hjmubah}(h]h ]h"]h$]h&]uh1hhjjubh)}(h``local`` - child namespaces will have private CID allocation and their sockets will only be able to connect within their own namespace. h]h)}(h``local`` - child namespaces will have private CID allocation and their sockets will only be able to connect within their own namespace.h](j)}(h ``local``h]hlocal}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh - child namespaces will have private CID allocation and their sockets will only be able to connect within their own namespace.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM$hjubah}(h]h ]h"]h$]h&]uh1hhjjubeh}(h]h ]h"]h$]h&]jUjVuh1hhhhM!hjfubah}(h]h ]h"]h$]h&]uh1hhhhM!hjhhubh)}(hThe first write to ``child_ns_mode`` locks its value. Subsequent writes of the same value succeed, but writing a different value returns ``-EBUSY``.h](hThe first write to }(hjhhhNhNubj)}(h``child_ns_mode``h]h child_ns_mode}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubhe locks its value. Subsequent writes of the same value succeed, but writing a different value returns }(hjhhhNhNubj)}(h ``-EBUSY``h]h-EBUSY}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM(hjhhubh)}(hChanging ``child_ns_mode`` only affects namespaces created after the change; it does not modify the current namespace or any existing children.h](h Changing }(hjhhhNhNubj)}(h``child_ns_mode``h]h child_ns_mode}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubhu only affects namespaces created after the change; it does not modify the current namespace or any existing children.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM+hjhhubh)}(hqA namespace with ``ns_mode`` set to ``local`` cannot change ``child_ns_mode`` to ``global`` (returns ``-EPERM``).h](hA namespace with }(hjhhhNhNubj)}(h ``ns_mode``h]hns_mode}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh set to }(hjhhhNhNubj)}(h ``local``h]hlocal}(hj1hhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh cannot change }(hjhhhNhNubj)}(h``child_ns_mode``h]h child_ns_mode}(hjChhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh to }(hjhhhNhNubj)}(h ``global``h]hglobal}(hjUhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh (returns }(hjhhhNhNubj)}(h ``-EPERM``h]h-EPERM}(hjghhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh).}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM.hjhhubeh}(h] child-ns-modeah ]h"] child_ns_modeah$]h&]uh1hhjhhhhhMubh)}(hhh](h)}(h g2h_fallbackh]h g2h_fallback}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhM2ubh)}(hControls whether connections to CIDs not owned by the host-to-guest (H2G) transport automatically fall back to the guest-to-host (G2H) transport.h]hControls whether connections to CIDs not owned by the host-to-guest (H2G) transport automatically fall back to the guest-to-host (G2H) transport.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM4hjhhubh)}(hXWhen enabled, if a connect targets a CID that the H2G transport (e.g. vhost-vsock) does not serve, or if no H2G transport is loaded at all, the connection is routed via the G2H transport (e.g. virtio-vsock) instead. This allows a host running both nested VMs (via vhost-vsock) and sibling VMs reachable through the hypervisor (e.g. Nitro Enclaves) to address both using a single CID space, without requiring applications to set ``VMADDR_FLAG_TO_HOST``.h](hXWhen enabled, if a connect targets a CID that the H2G transport (e.g. vhost-vsock) does not serve, or if no H2G transport is loaded at all, the connection is routed via the G2H transport (e.g. virtio-vsock) instead. This allows a host running both nested VMs (via vhost-vsock) and sibling VMs reachable through the hypervisor (e.g. Nitro Enclaves) to address both using a single CID space, without requiring applications to set }(hjhhhNhNubj)}(h``VMADDR_FLAG_TO_HOST``h]hVMADDR_FLAG_TO_HOST}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM7hjhhubh)}(hWhen the fallback is taken, ``VMADDR_FLAG_TO_HOST`` is automatically set on the remote address so that userspace can determine the path via ``getpeername()``.h](hWhen the fallback is taken, }(hjhhhNhNubj)}(h``VMADDR_FLAG_TO_HOST``h]hVMADDR_FLAG_TO_HOST}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubhY is automatically set on the remote address so that userspace can determine the path via }(hjhhhNhNubj)}(h``getpeername()``h]h getpeername()}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM?hjhhubh)}(hNote: With this sysctl enabled, user space that attempts to talk to a guest CID which is not implemented by the H2G transport will create host vsock traffic. Environments that rely on H2G-only isolation should set it to 0.h]hNote: With this sysctl enabled, user space that attempts to talk to a guest CID which is not implemented by the H2G transport will create host vsock traffic. Environments that rely on H2G-only isolation should set it to 0.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMChjhhubh)}(hValues:h]hValues:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMGhjhhubh)}(h- 0 - Connections to CIDs <= 2 or with VMADDR_FLAG_TO_HOST use G2H; all others use H2G (or fail with ENODEV if H2G is not loaded). - 1 - Connections to CIDs not owned by H2G fall back to G2H. (default)h]h)}(hhh](h)}(h0 - Connections to CIDs <= 2 or with VMADDR_FLAG_TO_HOST use G2H; all others use H2G (or fail with ENODEV if H2G is not loaded).h]h)}(h0 - Connections to CIDs <= 2 or with VMADDR_FLAG_TO_HOST use G2H; all others use H2G (or fail with ENODEV if H2G is not loaded).h]h0 - Connections to CIDs <= 2 or with VMADDR_FLAG_TO_HOST use G2H; all others use H2G (or fail with ENODEV if H2G is not loaded).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMIhjubah}(h]h ]h"]h$]h&]uh1hhjubh)}(hD1 - Connections to CIDs not owned by H2G fall back to G2H. (default)h]h)}(hj5h]hD1 - Connections to CIDs not owned by H2G fall back to G2H. (default)}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMKhj3ubah}(h]h ]h"]h$]h&]uh1hhjubeh}(h]h ]h"]h$]h&]jUjVuh1hhhhMIhjubah}(h]h ]h"]h$]h&]uh1hhhhMIhjhhubeh}(h] g2h-fallbackah ]h"] g2h_fallbackah$]h&]uh1hhjhhhhhM2ubeh}(h]&proc-sys-net-core-network-core-optionsah ]h"],1. /proc/sys/net/core - network core optionsah$]h&]uh1hhhhhhhhK/ubeh}(h]documentation-for-proc-sys-netah ]h"] documentation for /proc/sys/net/ah$]h&]uh1hhhhhhhhKubeh}(h]h ]h"]h$]h&]sourcehuh1hcurrent_sourceN current_lineNsettingsdocutils.frontendValues)}(hN generatorN datestampN source_linkN source_urlN toc_backlinksjfootnote_backlinksK sectnum_xformKstrip_commentsNstrip_elements_with_classesN strip_classesN report_levelK halt_levelKexit_status_levelKdebugNwarning_streamN tracebackinput_encoding utf-8-siginput_encoding_error_handlerstrictoutput_encodingutf-8output_encoding_error_handlerjerror_encodingutf-8error_encoding_error_handlerbackslashreplace language_codeenrecord_dependenciesNconfigN id_prefixhauto_id_prefixid dump_settingsNdump_internalsNdump_transformsNdump_pseudo_xmlNexpose_internalsNstrict_visitorN_disable_configN_sourcehʌ _destinationN _config_files]7/var/lib/git/docbuild/linux/Documentation/docutils.confafile_insertion_enabled raw_enabledKline_length_limitM'pep_referencesN pep_base_urlhttps://peps.python.org/pep_file_url_templatepep-%04drfc_referencesN rfc_base_url&https://datatracker.ietf.org/doc/html/ tab_widthKtrim_footnote_reference_spacesyntax_highlightlong smart_quotessmartquotes_locales]character_level_inline_markupdoctitle_xform docinfo_xformKsectsubtitle_xform image_loadinglinkembed_stylesheetcloak_email_addressessection_self_linkenvNubreporterNindirect_targets]substitution_defs}substitution_names}refnames}refids}nameids}(jkjhjcj`j7j4jjjJ jG jq jn j j j j j j jS jP j j j j j j j5 j2 j\ jY j j j j j j j j jI jF jp jm j j j j j j j( j% j] jZ j j j'j$jNjKjjjjjjj>j;jsjpjjjjj.j+jUjRj|jyjjjjjjjLjIjjjjj[jXu nametypes}(jkjcj7jjJ jq j j j jS j j j j5 j\ j j j j jI jp j j j j( j] j j'jNjjjj>jsjjj.jUj|jjjjLjjj[uh}(jhhj`jj4jjj:jG jjn jM j jt j j j j jP j! j jV j j j j j2 j jY j8 j j_ j j j j j j jF j jm jL j js j j j j j% j jZ j+ j j` j$j jKj*jjQjjjjj;jjpjAjjvjjj+jjRj1jyjXjjjjjjjIjjjOjjjXju footnote_refs} citation_refs} autofootnotes]autofootnote_refs]symbol_footnotes]symbol_footnote_refs] footnotes] citations]autofootnote_startKsymbol_footnote_startK id_counter collectionsCounter}Rparse_messages]transform_messages] transformerN include_log] decorationNhhub.