📄 changelog
字号:
commit 6647c90e6c44cfbd2188696e4ee3ad37284d8e50Author: Florian Lohoff <flo@rfc822.org>Date: Thu Dec 4 15:05:02 2008 +0100 Add more files to ignore for gitcommit 56d591d72e84dd6da3ab185ce9b24965304a9926Author: Romolo Manfredini <romolo@ms-informatica.it>Date: Thu Dec 4 10:32:03 2008 +0100 Rewrite DiseqC setup codecommit 05c581b9effa70f30b28420649a4b91677be2e89Author: Florian Lohoff <flo@rfc822.org>Date: Wed Sep 17 07:37:31 2008 +0200 Fix another VLC 0.9.2 incompatibility Since 0.9.2 vlc refused to accept a hex session id in SAP announcements.commit b8c48e15e108993051f87cd23fa6f0fa0d4a0027Author: Florian Lohoff <flo@rfc822.org>Date: Tue Sep 16 17:18:12 2008 +0200 Add a PSI sanity check Damjan Marion reported getstream crashes in case of a high BER which were caused by the pointer field beeing corrupted.commit 70a81959162b8dc2de688fa7275a6fefc751002aAuthor: Florian Lohoff <flo@rfc822.org>Date: Tue Sep 16 14:03:16 2008 +0200 Another bug hint from Egor Shibeko with the PSI versioningcommit 2598fd5f8d024f24a45395cc58adcd4ee9b2670cAuthor: Florian Lohoff <flo@rfc822.org>Date: Tue Sep 16 12:05:59 2008 +0200 Fix hang in PSI reassembly Got a report about a possible hang. It seems when a PSI section would exactly consume the full TS packet up to the last byte getstream could hang.commit 5a8c04d9678f10c60e03c0eb8cc810dce9f7d13aAuthor: Florian Lohoff <flo@rfc822.org>Date: Tue Sep 16 11:19:23 2008 +0200 Fix adaption field size calculation This did not occur for me as i dont see any adaption field but in case we have an adaption field the psi would not work.commit e2d242c94db84e54c8e061f25b5d7db29a2e8311Author: Florian Lohoff <flo@rfc822.org>Date: Tue Sep 16 11:08:22 2008 +0200 Fix SDP incompatibility with VLC 0.9.2 Change order of SDP lines in the SAP announcements. VLC 0.9.2 seems to be picky about ordering although i cant seem to find it in the RFC. Pull o= to the top.commit b1af6bce8d6d23647bfb0794ff7b68bd2a9bb328Author: Florian Lohoff <flo@rfc822.org>Date: Sat Apr 26 06:29:26 2008 +0000 Follow API changes of Multiproto (DVBFE_SET_DELSYS)commit dc763bf68578bf8a1376a53466d66be2962c3061Author: Florian Lohoff <flo@rfc822.org>Date: Sat Apr 26 06:17:28 2008 +0000 Add Multiproto build instructionscommit 3f03e00c16f2c44267a8fbceef730136f0cbc45aAuthor: Florian Lohoff <flo@rfc822.org>Date: Sat Mar 1 21:10:09 2008 +0000 Fix HTTP memory leak (didnt free the http connection), reuse evbuffer in libhttp.ccommit b4e7b83c0af5146dee617a3ce3aca54782371002Author: Florian Lohoff <flo@rfc822.org>Date: Sat Mar 1 20:43:18 2008 +0000 Cleanup some DVB capability checks Needed for multiproto which makes it a broken API.commit 4eefaa8f2887cfcf171844036e542e84e323725cAuthor: Florian Lohoff <flo@rfc822.org>Date: Sat Mar 1 17:04:13 2008 +0000 Fix http bug/crash This bug is is remote triggerable by sending data in the http connection which is not the right formatcommit 086d94229d6c14141a1c74d8cb7f730f25319f45Author: Florian Lohoff <flo@rfc822.org>Date: Sat Mar 1 15:09:16 2008 +0100 Fix Multiproto init bug With multiproto one HAS TO ISSUE an ioctl DVBFE_GET_INFO otherwise the information about the delivery system is not known in kernel - I think this is a kernel bug ...commit 94423d2c52fcd3a1bf51dc65a884f7467655ae72Author: Florian Lohoff <flo@rfc822.org>Date: Wed Feb 27 11:04:27 2008 +0100 Add some hints about DVB-S2 to Documentationcommit 2097ec13b146945b17c151eaaa1bb932521c9331Author: Florian Lohoff <flo@rfc822.org>Date: Wed Feb 27 11:01:00 2008 +0100 Fix error messages on adapter type mismatchcommit 69c2df82e18e9ced5d1b0c97e97f890376eb73b3Author: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 26 12:11:21 2008 +0100 Fix some non Multiproto compile errorscommit e68b166ce3b192ec7f21961bdd9ce52067508100Author: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 26 13:08:55 2008 +0000 Add Multiproto supportcommit a4427be6acf555e9dbddc325d3698a2c24918558Author: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 26 10:24:53 2008 +0000 Fix small error log bug in fe.c concerning FEC_AUTOcommit 90ad8b10258fd560b003e6b87932261db249f7ebAuthor: Florian Lohoff <flo@rfc822.org>Date: Thu Feb 21 08:27:19 2008 +0100 The frontend filedescriptor may actually be 0 Thank goes to Karel Tuma for the patch.commit 5f5c8b35a865454fd9ed1295b19555481370382fAuthor: Florian Lohoff <flo@rfc822.org>Date: Thu Feb 21 08:12:41 2008 +0100 Added setsockopt for the ttl to output_rtp.ccommit 626f690a94799c46a94388b17e99f4e4a17937d4Author: Florian Lohoff <flo@rfc822.org>Date: Thu Feb 21 08:07:25 2008 +0100 Documentation for the input section e.h. pid, pnr and fullcommit 31a971139fe8c2e62be003fe0fd27a62ddcb4a69Author: Florian Lohoff <flo@rfc822.org>Date: Thu Feb 21 07:41:00 2008 +0100 More error msg on send error in output_udp, Add full transpondercommit 9cfcde6534419c718d953b2e3c87f18ee553cce9Author: Florian Lohoff <flo@rfc822.org>Date: Wed Feb 20 12:43:15 2008 +0100 Fix local-address beeing mandatory for udp and rtpcommit d6adb306c96b37cdf08f8c376cc9760666cdd17dAuthor: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 19 19:17:30 2008 +0100 Add possibility to bind to local address for output_{udp,rtp}commit af56a42d016f3aa1a6491199849eaaec81b8251fAuthor: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 19 18:57:58 2008 +0100 Fix RTP output SAP config - was removed on getstream2 rewritecommit 61a0a29faa22dbfc11a0d4a4741645bbc7cebde5Author: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 19 18:33:05 2008 +0100 Fix error paths in output_{udp,rtp}.c and add socket_closecommit c40b84599ffb8f896c850c114e21e205246c38fbAuthor: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 19 15:57:44 2008 +0100 Convert SAP to socket helpers and make it join the MCast groupcommit 70d55fef07e25f232f8b025b0124194f68244353Author: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 19 15:46:57 2008 +0100 Fix a socket_set_nonblock in output_rtp.ccommit 15f2a3ae3cd6120c782d50d2d074b4ead7866dd4Author: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 19 15:44:23 2008 +0100 Fix startup crash in PSI reassembly Fix an startup crash caused by trying to continue a non started psi reassembly. Use the opportunity to simplify the decisions. Backtrace looked like this: #0 0x0804b351 in crc32_be (crc=3072780846, p=0x806ee90 "", len=-192881) at crc32.c:413 #1 0x08051406 in psi_ccrc (section=0x806ee80) at psi.c:25 #2 0x08051427 in psi_crc_valid (section=0x806ee80) at psi.c:30 #3 0x0805179c in psi_reassemble (section=0x806ee80, ts=0x80602c8 "G", off=0) at psi.c:178 #4 0x0804fb93 in dvr_input_ts (a=0xb7cd5008, ts=0x80602c8 "G") at dvr.c:64 #5 0x0804fa63 in dvr_read (fd=8, event=2, arg=0xb7cd5008) at dvr.c:168 #6 0xb7eeac79 in event_base_priority_init () from /usr/lib/libevent-1.1a.so.1 #7 0xb7eeaf65 in event_base_loop () from /usr/lib/libevent-1.1a.so.1 #8 0xb7eeadcb in event_loop () from /usr/lib/libevent-1.1a.so.1 #9 0xb7eeacb0 in event_dispatch () from /usr/lib/libevent-1.1a.so.1 #10 0x08049d42 in main (argc=4, argv=0xbfad8d74) at getstream.c:177commit a8b801f854fb4e403025cbc5165286737cfbcb46Author: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 19 14:43:29 2008 +0100 Fix endianess Bug in IN_MULTICASTcommit 0b7ddc67bf4aa03def01d54c4298ad10ebbeb906Author: Florian Lohoff <flo@rfc822.org>Date: Tue Feb 19 15:37:55 2008 +0100 Add some socket helpers, drop RTCP, add MCast joins * Add socket helpers * Convert output_rtp and output_udp to socket helpers * Drop RTCP implementations (#if 0) because its incomplete * Add MULTICAST joins to output_udp and output_rtpcommit a33a6cb51487f3ec483c82f28f41fa589de4b825Author: Florian Lohoff <flo@rfc822.org>Date: Mon Oct 29 14:17:44 2007 +0100 Make lnb-sharing optional and add stuck-interval check optioncommit 83f60c31bae813d279b700412a6a3c157664cd48Author: Florian Lohoff <flo@rfc822.org>Date: Thu Oct 25 10:02:53 2007 +0200 Add lnb-sharing documentationcommit 910a8b71fd0d15b85a4f064a2336bfc5074338caAuthor: Florian Lohoff <flo@rfc822.org>Date: Thu Oct 25 09:55:22 2007 +0200 Add DVB-S config and stat-interval to READMEcommit 706e5eb9a3c4e7f76f2b4091d1146ab13caa9df2Author: Florian Lohoff <flo@rfc822.org>Date: Sun Sep 23 17:27:17 2007 +0200 Add lnb sharing option Add lnb-sharing option to let the frontend code turn off lnb influencing tones and voltages.commit 843320225720e743321896a204f76f6c52881f73Author: Florian Lohoff <flo@rfc822.org>Date: Sun Sep 23 10:13:26 2007 +0200 Check for correct card capabilities and dump fe setup Add FE_GET_INFO and check for correct card type and Dump frontend capabilities for debuggingcommit 421a63335e40d21e2e8c6e120bd38af0e9cd8d17Author: Florian Lohoff <flo@rfc822.org>Date: Sat Sep 15 12:10:40 2007 +0200 Fix another stuck card issue Fix another stuck card issue. Sometimes not only closing and reopening the filters is needed but also retuning.commit 292dc295a2c53b1206122fb70ae04cf4da15d63fAuthor: Florian Lohoff <flo@rfc822.org>Date: Tue Sep 4 15:05:11 2007 +0200 Fix incrementing read counter. Fix filter bouncing * Fix a flaw in not incrementing the read counter * Stop/Start the demux filter with DMX_STOP/DMX_START ioctlcommit 5211636c3ae08ba2c0f4205baa89c0fc2d3b99a3Author: Florian Lohoff <flo@rfc822.org>Date: Tue Sep 4 13:14:27 2007 +0200 Clean up more object filescommit 46f8b02a1390c6068232d475e480d3dda9ef1614Author: Florian Lohoff <flo@rfc822.org>Date: Tue Sep 4 13:13:21 2007 +0200 Fix namespace. Remove PAT special case, add bounce filter * Fixup some namespaces in getstream.h/struct adapter_s * Remove PAT special case where we let the kernel reassemble the PAT section. We have our own PSI section reassembler so let it do its job * Add a function dmx_bounce_filter which should remove all PID filters and thus leave all PIDs and rejoin them which should reset the flexcop. Retuning in the patches sense means - close everything - change transponder, join all pids. From conversation with Patrick Boettcher the reset happens on the transition from forward pid count 0 to 1.commit fe379584b499d0f7058bf68a59f9a0cff27bc639Author: Florian Lohoff <flo@rfc822.org>Date: Mon Sep 3 12:39:55 2007 +0200 Add flexcop/SkyStar2 IRQ Stop workaround * Rewrote frontend handling without a thread. The ioctls should not block too long to do it in a poll/select loop. * Added flexcop aka SkyStar2 IRQ Stop workaround. The Kernel workaround is not enough as it only resets the card in case of a retune. As we never retune it will lockup one day. We now retune in case there are no more packets from the dvr device. We might try retuning regularly in case we still get packets but i am unshure on the effect on other cards.commit f15f0a62a75e3ed4990cca90ed427ed0bdbf39a7Author: Florian Lohoff <flo@rfc822.org>Date: Thu Dec 4 14:17:47 2008 +0100 Initial import
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -