aboutsummaryrefslogtreecommitdiff
path: root/networking
AgeCommit message (Collapse)Author
2018-06-24tls: fix to handle X.509 v1 certificates correctlyIvan Abrea
The syntax of public key certificates can be found in RFC 5280 section 4.1. The relevant part of the syntax is the following: TBSCertificate ::= SEQUENCE { version [0] EXPLICIT Version DEFAULT v1, serialNumber CertificateSerialNumber, ... remaining fields omitted ... } The version field has a default value of v1. RFC 5280 section 4.1.2.1 says the following: If only basic fields are present, the version SHOULD be 1 (the value is omitted from the certificate as the default value); however, the version MAY be 2 or 3. To help detect if the version field is present or not, the type of the version field has an explicit tag of [0]. Due to this tag, if the version field is present, its encoding will have an identifier octet that is distinct from that of the serialNumber field. ITU-T X.690 specifies how a value of such a type should be encoded with DER. There is a PDF of X.690 freely available from ITU-T. X.690 section 8.1.2 specifies the format of identifier octets which is the first component of every encoded value. Identifier octets encode the tag of a type. Bits 8 and 7 encode the tag class. Bit 6 will be 0 if the encoding is primitive and 1 if the encoding is constructed. Bits 5 to 1 encode the tag number. X.690 section 8.14 specifies what the identifier octet should be for explicitly tagged types. Section 8.14.3 says if implicit tagging is not used, then the encoding shall be constructed. The version field uses explicit tagging and not implicit tagging, so its encoding will be constructed. This means bit 6 of the identifier octet should be 1. X.690 section 8.14 and Annex A provide examples. Note from their examples that the notation for tags could look like [APPLICATION 2] where both the tag class and tag number are given. For this example, the tag class is 1 (application) and the tag number is 2. For notation like [0] where the tag class is omitted and only the tag number is given, the tag class will be context-specific. Putting this all together, the identifier octet for the DER encoding of the version field should have a tag class of 2 (context-specific), bit 6 as 1 (constructed), and a tag number of 0. Signed-off-by: Ivan Abrea <ivan@algosolutions.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-06-24ntpd: deprecate IPTOS_LOWDELAY in favor of IPTOS_DSCP_AF21Codarren Velvindron
Update QoS markers. Use DSCP AF21 for interactive traffic. DSCP is defined in RFC2474. Many modern equipment no longer support IPTOS. Signed-off-by: Codarren Velvindron <codarren@hackers.mu> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-06-21udhcpc: remove code which requires server ID to be on local networkDenys Vlasenko
This reverts "udhcpc: paranoia when using kernel UDP mode for sending renew: server ID may be bogus". Users complain that they do have servers behind routers (with DHCP relays). function old new delta send_packet 168 166 -2 bcast_or_ucast 25 23 -2 udhcp_send_kernel_packet 301 295 -6 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-10) Total: -10 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-06-06restore documentation on the build config languageKartik Agaram
Kconfig-language.txt was deleted in commit 4fa499a17b52b back in 2006. Move to docs/ as suggested by Xabier Oneca: http://lists.busybox.net/pipermail/busybox/2014-May/080914.html Also update references to it everywhere. Signed-off-by: Kartik Agaram <akkartik@gmail.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-28wget: emit a message that certificate verification is not implementedDenys Vlasenko
function old new delta spawn_ssl_client 185 209 +24 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-24udhcpc6: carry along length of packet when parsing it.David Decotigny
This is to avoid parsing garbage past packet's actual end. Also const-ize params to a few functions. function old new delta d6_run_script_no_option - 12 +12 option_to_env 791 798 +7 d6_run_script 253 255 +2 perform_d6_release 95 93 -2 udhcpc6_main 2596 2592 -4 ------------------------------------------------------------------------------ (add/remove: 1/0 grow/shrink: 2/2 up/down: 21/-6) Total: 15 bytes Signed-off-by: David Decotigny <ddecotig@gmail.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-24udhcpc6: ignore invalid OPTION_IAADDRDavid Decotigny
Signed-off-by: David Decotigny <ddecotig@gmail.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-24nc: fix the !NC_SERVER configurationDenys Vlasenko
The symptom is the "nc: NO OPT l!" message. Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-14udhcpc6: set -x options in requestDenys Vlasenko
Last foru commits: function old new delta option_to_env 621 791 +170 .rodata 168351 168505 +154 attach_option 431 506 +75 add_d6_client_options 112 167 +55 d6_option_strings 30 84 +54 udhcp_str2optset 644 660 +16 d6_optflags 12 20 +8 udhcpc6_main 2590 2596 +6 udhcpc_main 2648 2651 +3 read_optset 15 18 +3 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 10/0 up/down: 544/0) Total: 544 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-14udhcpc6: add PXELINUX optionsSamuel Mendoza-Jonas
Add support for the PXELINUX options 209 ("ConfigFile") and 210 ("PathPrefix") in the DHCPv6 client. See also: RFC5071, "Dynamic Host Configuration Protocol Options Used by PXELINUX". Signed-off-by: Samuel Mendoza-Jonas <sam@mendozajonas.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-14udhcpc6: add DHCPv6 boot optionsSamuel Mendoza-Jonas
Add support for 'bootfile-url' and 'bootfile-params' as defined by RFC5970 "DHCPv6 Options for Network Boot". Signed-off-by: Samuel Mendoza-Jonas <sam@mendozajonas.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-14udhcpc6: add DHCPv6 env helperSamuel Mendoza-Jonas
Add string_option_to_env() to easily generate environment variables for known simple options. Signed-off-by: Samuel Mendoza-Jonas <sam@mendozajonas.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-05-14udhcp: fix use of trim() in udhcp_str2optset()Samuel Mendoza-Jonas
trim() modifies the string in place if needed and returns a pointer to the end of the resulting string. Update udhcp_str2optset() so it no longer sets the value of 'val' to the return value of trim(). Signed-off-by: Samuel Mendoza-Jonas <sam@mendozajonas.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-29ifplugd: close signal raceDenys Vlasenko
function old new delta ifplugd_main 1109 1117 +8 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-24wget: handle URLs with @ or hash differentlyDenys Vlasenko
If server replied with 302 and Location: ?foo we used to underflow the allocated space while trying to form the "@foo" filename. Switch to forming "foo" filename. function old new delta packed_usage 32795 32799 +4 parse_url 387 352 -35 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 1/1 up/down: 4/-35) Total: -31 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-17libbb: new option FEATURE_ETC_SERVICES: if off, /etc/services reads often ↵Denys Vlasenko
avoided In practice, "wget http://host.com/" always uses port 80. People explicitly set non-standard ports via options or parameters ("telnet 1.2.3.4 567" or "telnet 1.2.3.4 ftp") instead of modifying /etc/services. function old new delta telnet_main 1466 1464 -2 rdate_main 215 198 -17 fakeidentd_main 269 252 -17 parse_url 459 392 -67 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 0/4 up/down: 0/-103) Total: -103 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-15nslookup: simplify make_ptrDenys Vlasenko
function old new delta nslookup_main 2644 2611 -33 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-15nslookup: get rid of query::rlen fieldDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-15nslookup: move array of queries to "globals"Denys Vlasenko
function old new delta add_query 95 89 -6 nslookup_main 2692 2641 -51 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-57) Total: -57 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-15nslookup: accept lowercase -type=soa, document query typesDenys Vlasenko
Usage: nslookup [-type=QUERY_TYPE] [-debug] HOST [DNS_SERVER] Query DNS about HOST QUERY_TYPE: soa,ns,a,aaaa,cname,mx,txt,ptr,any function old new delta packed_usage 32189 32258 +69 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-15nslookup: change -stats to -debug (it's a bug in bind that it accepts -s)Denys Vlasenko
function old new delta packed_usage 32211 32189 -22 nslookup_main 2754 2692 -62 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-84) Total: -84 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-15nslookup: rework option parsingDenys Vlasenko
function old new delta nslookup_main 2715 2754 +39 packed_usage 32179 32211 +32 add_ns 65 66 +1 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 3/0 up/down: 72/0) Total: 72 bytes text data bss dec hex filename 926262 555 5740 932557 e3acd busybox_old 926239 555 5740 932534 e3ab6 busybox_unstripped Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-15nslookup: smaller qtypes[] arrayDenys Vlasenko
function old new delta nslookup_main 2708 2715 +7 qtypes 80 72 -8 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 1/1 up/down: 7/-8) Total: -1 bytes text data bss dec hex filename 926277 555 5740 932572 e3adc busybox_old 926262 555 5740 932557 e3acd busybox_unstripped Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-15nslookup: process replies immediately, do not store themDenys Vlasenko
function old new delta nslookup_main 1837 2708 +871 parse_reply 852 - -852 ------------------------------------------------------------------------------ (add/remove: 0/1 grow/shrink: 1/0 up/down: 871/-852) Total: 19 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-14nslookup: fix output corruption for "nslookup 1.2.3.4"Denys Vlasenko
function old new delta nslookup_main 1832 1837 +5 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-14nslookup: more closely resemble output format of bind-utils-9.11.3Denys Vlasenko
function old new delta nslookup_main 1880 1832 -48 parse_reply 1022 852 -170 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-218) Total: -218 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-14nslookup: shrink send_queries()Denys Vlasenko
function old new delta rcodes 68 64 -4 nslookup_main 2007 1880 -127 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-131) Total: -131 bytes text data bss dec hex filename 926735 555 5740 933030 e3ca6 busybox_old 926525 555 5740 932820 e3bd4 busybox_unstripped Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-14nslookup: use xmalloc_sockaddr2dotted() instead of homegrown functionDenys Vlasenko
function old new delta nslookup_main 2091 2007 -84 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-14nslookup: usee bbox network functions instead of opne-coded messDenys Vlasenko
function old new delta nslookup_main 2363 2091 -272 add_ns 663 65 -598 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-870) Total: -870 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-14nslookup: add openwrt / lede versionDenys Vlasenko
Needs work on size reduction function old new delta nslookup_main 114 2363 +2249 parse_reply - 1022 +1022 add_ns - 663 +663 ns_parserr - 486 +486 ns_initparse - 184 +184 ns_skiprr - 117 +117 add_query - 95 +95 qtypes - 80 +80 rcodes - 68 +68 dn_skipname - 58 +58 ns_name_uncompress - 56 +56 ns_get16 - 13 +13 v4_mapped - 12 +12 ns_get32 - 9 +9 res_init 3 - -3 __res_state 6 - -6 xmalloc_sockaddr2hostonly_noport 10 - -10 gai_strerror 47 - -47 set_default_dns 95 - -95 print_host 199 - -199 static.res 512 - -512 ------------------------------------------------------------------------------ (add/remove: 15/10 grow/shrink: 1/0 up/down: 5112/-872) Total: 4240 bytes text data bss dec hex filename 921944 555 6252 928751 e2bef busybox_old 927375 555 5740 933670 e3f26 busybox_unstripped Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-13udhcp: support string user options, closes 10946Denys Vlasenko
function old new delta udhcp_str2optset 536 628 +92 packed_usage 32757 32760 +3 udhcpc_main 2708 2692 -16 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 2/1 up/down: 95/-16) Total: 79 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-08wget: check chunk length for overflowing off_tDenys Vlasenko
function old new delta retrieve_file_data 428 465 +37 wget_main 2386 2389 +3 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 2/0 up/down: 40/0) Total: 40 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-08libbb.h: always include sys/resource.hDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-07telnet: move winsize detection closer to I/O loop, delete non-functioning ↵Denys Vlasenko
debug code Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-07ftpd: allow -A if !FTPD_AUTHENTICATION as wellDenys Vlasenko
Users will be able to use "ftpd -A" in scripts regardless of build config Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-07ftpd: added -A option to disable all authentication, closes 10921Denys Vlasenko
function old new delta packed_usage 32745 32777 +32 ftpd_main 2162 2156 -6 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 1/1 up/down: 32/-6) Total: 26 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-07wget,ftpd: shorten and reuse stringsDenys Vlasenko
function old new delta wget_main 2382 2386 +4 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 1/0 up/down: 4/0) Total: 4 bytes text data bss dec hex filename 934228 477 7296 942001 e5fb1 busybox_old 934202 477 7296 941975 e5f97 busybox_unstripped Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-07httpd: do not default to Content-type: application/octet-streamDenys Vlasenko
Instead, simply don't send this header. On Mon, Apr 2, 2018 at 8:17 PM, xisd <xisd-dev@riseup.net> wrote: > I had some trouble using busybox httpd to serve a static website and I > thought the issue might be of interest. > > My problem is related to something that seem quite common for static > site generator : the use of html files without the '.html' extension > (it is called 'clean url'...) > > Most web server guess that these files are html and display them like > any other .html files. > > From what I understood, the MIME type for files without extension in > busybox htttp default settings is 'application/octet-stream', and > because of that 'clean url' pages are not displayed. > > It is only trouble because I wanted to deploy my website on freshly > installed linux without editing any configuration. > > The default MIME setting make sense to me as it is, I just thought that > might be worth mentioning since the use of 'clean url' seem to be a > common practice for static sites generators (the one I use is callled > 'yellow' (https://github.com/datenstrom/yellow)) > > Here is a link for the related issue on github : > https://github.com/datenstrom/yellow/issues/317 function old new delta send_headers 702 718 +16 send_headers_and_exit 23 20 -3 handle_incoming_and_exit 2794 2791 -3 send_file_and_exit 772 756 -16 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 1/3 up/down: 16/-22) Total: -6 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-04-01libbb: new function bb_die_memory_exhaustedDenys Vlasenko
function old new delta bb_die_memory_exhausted - 10 +10 xstrdup 28 23 -5 xsetenv 27 22 -5 xrealloc 32 27 -5 xputenv 22 17 -5 xmalloc 30 25 -5 xfdopen_helper 40 35 -5 xasprintf 44 39 -5 wget_main 2387 2382 -5 open_socket 54 49 -5 glob_brace 419 414 -5 bb_get_chunk_from_file 146 141 -5 ------------------------------------------------------------------------------ (add/remove: 1/0 grow/shrink: 0/11 up/down: 10/-55) Total: -45 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-29ifplugd: if SIOCSIFFLAGS fails with EADDRNOTAVAIL, don't dieStefan Agner
When using ifplugd on a RNDIS interface with monitor mode, using SIOCSIFFLAGS seems to fail with EADDRNOTAVAIL. Treat it the same as if iface disappeared again. This has been observed on a Tegra TK1 using Linux 4.14. There seem to be a race when the kernel is creating the device: ifplugd(usb0): started: BusyBox v1.24.1 (2018-03-27 09:47:48 CEST) ifplugd(usb0): interface doesn't exist, waiting ifplugd(usb0): interface appeared ifplugd(usb0): upping interface ifplugd(usb0): setting interface flags failed: Cannot assign requested address <exit> With this patch ifplugd is successful in upping the device the second time around: ifplugd(usb0): upping interface ifplugd(usb0): using SIOCETHTOOL detection mode ifplugd(usb0): interface appeared ifplugd(usb0): upping interface ifplugd(usb0): setting interface flags failed: Cannot assign requested address ifplugd(usb0): interface appeared ifplugd(usb0): upping interface ifplugd(usb0): using SIOCETHTOOL detection mode ifplugd(usb0): link is up ifplugd(usb0): executing '/etc/ifplugd/ifplugd.usb.action usb0 up' ifplugd up ifplugd(usb0): exit code: 0 function old new delta up_iface 120 127 +7 Signed-off-by: Stefan Agner <stefan@agner.ch> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-28ifupdown: do not fail if interface disappears during ifdownKaarle Ritvanen
Interface may not exist because it got deleted by an ifdown hook script earlier. This may happen when a virtual interface, such as VLAN, has multiple iface blocks defined. function old new delta static_down6 14 40 +26 static_down 54 70 +16 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 2/0 up/down: 42/0) Total: 42 bytes Signed-off-by: Kaarle Ritvanen <kaarle.ritvanen@datakunkku.fi> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-27ioctl(SIOCGIFINDEX) does not require clearing of entire ifrDenys Vlasenko
function old new delta INET6_setroute 492 472 -20 do_iplink 1357 1330 -27 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-47) Total: -47 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-27netstat: produce numeric-ip output for non-resolved namesMark Marshall
If we don't ask for numerical output, and the symbolic look-up failed we used to get "(null)", but the numeric output would be better. function old new delta ip_port_str 109 121 +12 Signed-off-by: Mark Marshall <mark.marshall@omicronenergy.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-20ssl_client: fix option parsingRon Yorston
The wrong character was used to indicate options taking an integer parameter. Signed-off-by: Ron Yorston <rmy@pobox.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-11tcpsvd: fix fallout from opt_complementary removalDenys Vlasenko
text data bss dec hex filename 933035 473 6836 940344 e5938 busybox_old 933051 473 6836 940360 e5948 busybox_unstripped Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-11ntpd: choose initial FREQHOLD_cnt dynamicallyDenys Vlasenko
function old new delta update_local_clock 834 858 +24 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-11udhcpd: clamp down huge auto_times to ~2M seconds, better EINTR poll handlingDenys Vlasenko
EINTR _should_ only happen on two signals we trap, and safe_poll _should_ work here just fine, but there were kernel bugs where spurious EINTRs happen (e.g. on ptrace attach). Be safe. function old new delta udhcpd_main 1437 1468 +31 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-10ntpd: suppress in-kernel frequency correction in first 8 adjtimex callsDenys Vlasenko
In other words: try to correct initially existing clock offset first, before assuming that our clock drifts. function old new delta update_local_clock 826 834 +8 ntp_init 550 557 +7 filter_datapoints 179 173 -6 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 2/1 up/down: 15/-6) Total: 9 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-10udhcpd: fix "not dying on SIGTERM"Denys Vlasenko
Fixes: commit 52a515d18724bbb34e3ccbbb0218efcc4eccc0a8 "udhcp: use poll() instead of select()" Feb 16 2017 udhcp_sp_read() is meant to check whether signal pipe indeed has some data to read. In the above commit, it was changed as follows: - if (!FD_ISSET(signal_pipe.rd, rfds)) + if (!pfds[0].revents) return 0; The problem is, the check was working for select() purely by accident. Caught signal interrupts select()/poll() syscalls, they return with EINTR (regardless of SA_RESTART flag in sigaction). _Then_ signal handler is invoked. IOW: they can't see any changes to fd state caused by signal haldler (in our case, signal handler makes signal pipe ready to be read). For select(), it means that rfds[] bit array is unmodified, bit of signal pipe's read fd is still set, and the above check "works": it thinks select() says there is data to read. This accident does not work for poll(): .revents stays clear, and we do not try reading signal pipe as we should. In udhcpd, we fall through and block in socket read. Further SIGTERM signals simply cause socket read to be interrupted and then restarted (since SIGTERM handler has SA_RESTART=1). Fixing this as follows: remove the check altogether. Set signal pipe read fd to nonblocking mode. Always read it in udhcp_sp_read(). If read fails, assume it's EAGAIN and return 0 ("no signal seen"). udhcpd avoids reading signal pipe on every recvd packet by looping if EINTR (using safe_poll()) - thus ensuring we have correct .revents for all fds - and calling udhcp_sp_read() only if pfds[0].revents!=0. udhcpc performs much fewer reads (typically it sleeps >99.999% of the time), there is no need to optimize it: can call udhcp_sp_read() after each poll unconditionally. To robustify socket reads, unconditionally set pfds[1].revents=0 in udhcp_sp_fd_set() (which is before poll), and check it before reading network socket in udhcpd. TODO: This might still fail: if pfds[1].revents=POLLIN, socket read may still block. There are rare cases when select/poll indicates that data can be read, but then actual read still blocks (one such case is UDP packets with wrong checksum). General advise is, if you use a poll/select loop, keep all your fds nonblocking. Maybe we should also do that to our network sockets? function old new delta udhcp_sp_setup 55 65 +10 udhcp_sp_fd_set 54 60 +6 udhcp_sp_read 46 36 -10 udhcpd_main 1451 1437 -14 udhcpc_main 2723 2708 -15 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 2/3 up/down: 16/-39) Total: -23 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-03-08remove stray newline in "iplink --help"Denys Vlasenko
function old new delta packed_usage 32547 32550 +3 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>