------------------------------------------------------------------------ r1148 | sangorrin | 2008-05-03 12:11:17 +0200 (Sat, 03 May 2008) | 2 lines added Changelog file ------------------------------------------------------------------------ r1147 | sangorrin | 2008-05-03 12:10:17 +0200 (Sat, 03 May 2008) | 1 line added the SRC param for the fragmentation layer and a INSTALL file ------------------------------------------------------------------------ r1138 | sangorrin | 2008-04-23 18:09:46 +0200 (Wed, 23 Apr 2008) | 1 line found the bug.. actually it was a very serious bug that affected fragmentation of messages.. i forgot to change in a function that frag_id==0 is not for fp messages anymore.. now frag_id==MX_IDS is for fp messages, so we can use the rest 0..MX_IDS-1 for the servers ------------------------------------------------------------------------ r1137 | sangorrin | 2008-04-23 11:23:29 +0200 (Wed, 23 Apr 2008) | 1 line added the functions to renegotiate and cancel a contract. TODO: there is a bug when cancelling contracts, reneg async, reserve-commit, spare capacity, measurements ------------------------------------------------------------------------ r1136 | sangorrin | 2008-04-22 13:10:30 +0200 (Tue, 22 Apr 2008) | 1 line make a simplified utilization analysis for the moment ------------------------------------------------------------------------ r1135 | sangorrin | 2008-04-22 11:27:04 +0200 (Tue, 22 Apr 2008) | 2 lines add the infraestructure for storing the negotiated contracts and performing the analysis through the analysis module. TODO: perform a test with those contracts to admit them. In the end the module implementation could be replaced by the cpu implementation. Spare capacity is not done yet. Now we have to implement negotiation and cancel of contracts. ------------------------------------------------------------------------ r1131 | sangorrin | 2008-04-21 19:35:20 +0200 (Mon, 21 Apr 2008) | 1 line michael found an error in the definition of the posix sporadic servers so i replaced them with new rules designed by michael... I still keep them because they are still analyzable (they dont behave like periodic tasks though) to compare them ------------------------------------------------------------------------ r1124 | sangorrin | 2008-04-18 12:50:20 +0200 (Fri, 18 Apr 2008) | 1 line corrected an error in the sporadic servers... the activation time is the timestamp when the packet is enqueued, not when the frame is sent!!! ------------------------------------------------------------------------ r1123 | sangorrin | 2008-04-18 10:50:18 +0200 (Fri, 18 Apr 2008) | 1 line tested negotiation in slave and in master.. it seems to work good with the new structures too. And its much simpler and clear! wohooo!! TODO: do the scheduling analysis with the API that is already done but for the moment use a utilization analysis, then allow renegotiations and cancelling of contracts... finally implement modifications to the posix sporadic server... do testing, measures and the deliverable... (the spare capacity will have to wait i guess..) ------------------------------------------------------------------------ r1122 | sangorrin | 2008-04-17 20:54:04 +0200 (Thu, 17 Apr 2008) | 1 line some reorganization... now the acceptor is simpler.. everything goes through requests and these are sent and received directly through the messages module... TODO: sched analysis, test negotiations, then add renegotiations and cancel.. ------------------------------------------------------------------------ r1121 | sangorrin | 2008-04-17 17:52:19 +0200 (Thu, 17 Apr 2008) | 1 line rename the file ------------------------------------------------------------------------ r1120 | sangorrin | 2008-04-17 17:51:29 +0200 (Thu, 17 Apr 2008) | 1 line tests adapted to new names.. now the compile.. in makefile i add a target 'tests' to build them all in one shot ------------------------------------------------------------------------ r1119 | sangorrin | 2008-04-17 17:31:36 +0200 (Thu, 17 Apr 2008) | 1 line updated everything to the new structure... it compiles ok... next step, make sure that the tests compile, then check that the corresponding messages are sent and find possible bugs (very probable to have forgotten some info in the requests..) and then adapt it to do sched analysis ------------------------------------------------------------------------ r1118 | sangorrin | 2008-04-17 12:49:47 +0200 (Thu, 17 Apr 2008) | 1 line renamings... redo the request and messages part... also now there will be two threads in each node like in the dtm... still not compile.. i have to finish it in the afternoon ------------------------------------------------------------------------ r1117 | sangorrin | 2008-04-16 17:55:55 +0200 (Wed, 16 Apr 2008) | 2 lines renaming ------------------------------------------------------------------------ r1116 | sangorrin | 2008-04-16 17:27:03 +0200 (Wed, 16 Apr 2008) | 1 line i break the svn for one day to change the name of files, redo the negotiation messages system and implement renegotiations and canceling of contracts ------------------------------------------------------------------------ r1115 | sangorrin | 2008-04-15 13:47:40 +0200 (Tue, 15 Apr 2008) | 1 line update values ------------------------------------------------------------------------ r1114 | sangorrin | 2008-04-15 13:42:59 +0200 (Tue, 15 Apr 2008) | 1 line add a lock for the queue structures when dequeued, for th moment is hardwired to network 0 ------------------------------------------------------------------------ r1113 | sangorrin | 2008-04-15 13:42:29 +0200 (Tue, 15 Apr 2008) | 1 line this lock was wrong.. it worked because of an error in marte os: if u block after enter in the critical section, the idle task enters unblocking the irqs again. An error should be returned i think ------------------------------------------------------------------------ r1112 | sangorrin | 2008-04-15 10:51:36 +0200 (Tue, 15 Apr 2008) | 1 line lock for the (future) negotiation ------------------------------------------------------------------------ r1111 | sangorrin | 2008-04-15 10:51:12 +0200 (Tue, 15 Apr 2008) | 1 line net ------------------------------------------------------------------------ r1110 | sangorrin | 2008-04-15 10:50:57 +0200 (Tue, 15 Apr 2008) | 1 line locks for server functions ------------------------------------------------------------------------ r1109 | sangorrin | 2008-04-15 10:20:17 +0200 (Tue, 15 Apr 2008) | 1 line add the modules for the frescan sched analysis with an initial API that we are discussing at the moment so it might change soon... it is not applied yet.. TODO: we will need new messages for the spare capacity distribution and structures to apply them at the right moment.. this is also ongoing work and research must be done on when is the right moment. ------------------------------------------------------------------------ r1108 | sangorrin | 2008-04-15 09:47:57 +0200 (Tue, 15 Apr 2008) | 2 lines update the names and add acceptor thread id to the general data ------------------------------------------------------------------------ r1107 | sangorrin | 2008-04-15 09:44:38 +0200 (Tue, 15 Apr 2008) | 2 lines renaming to neg threads ------------------------------------------------------------------------ r1106 | sangorrin | 2008-04-15 09:43:54 +0200 (Tue, 15 Apr 2008) | 2 lines joining threads module ------------------------------------------------------------------------ r1105 | sangorrin | 2008-04-14 18:58:45 +0200 (Mon, 14 Apr 2008) | 1 line abort when higher prio frame ------------------------------------------------------------------------ r1104 | sangorrin | 2008-04-09 13:19:45 +0200 (Wed, 09 Apr 2008) | 1 line add a test to obtain measures of a round trip ------------------------------------------------------------------------ r1103 | sangorrin | 2008-04-09 13:19:14 +0200 (Wed, 09 Apr 2008) | 1 line apps need to link to libfosa too ------------------------------------------------------------------------ r1098 | sangorrin | 2008-04-08 17:37:53 +0200 (Tue, 08 Apr 2008) | 1 line network enabling is now in fna_configuration.h ------------------------------------------------------------------------ r1097 | sangorrin | 2008-04-08 17:37:00 +0200 (Tue, 08 Apr 2008) | 1 line initialize pointer to null to avoid warning.. this is because the compiler does not know that im checking that a list is not empty before using it etc.. ------------------------------------------------------------------------ r1096 | sangorrin | 2008-04-08 17:36:52 +0200 (Tue, 08 Apr 2008) | 1 line initialize pointer to null to avoid warning.. this is because the compiler does not know that im checking that a list is not empty before using it etc.. ------------------------------------------------------------------------ r1095 | sangorrin | 2008-04-08 17:36:42 +0200 (Tue, 08 Apr 2008) | 1 line initialize pointer to null to avoid warning.. this is because the compiler does not know that im checking that a list is not empty before using it etc.. ------------------------------------------------------------------------ r1094 | sangorrin | 2008-04-08 17:35:56 +0200 (Tue, 08 Apr 2008) | 1 line create a server in the negotiator thread... this thread should be renamed to MASTER local negotiator thread or sth like that cause the rest of negotiations are performed by the acceptor thread.. which also should be renamed to SLAVE negotiations or sth like that ------------------------------------------------------------------------ r1093 | sangorrin | 2008-04-08 17:34:51 +0200 (Tue, 08 Apr 2008) | 1 line implement send and receive in fna and bytes<->budget conversion functions... the test is in frsh/tests_distributed/test send receive basic ------------------------------------------------------------------------ r1092 | sangorrin | 2008-04-08 17:33:46 +0200 (Tue, 08 Apr 2008) | 1 line put neg channel to 0 by default so i dont forget it ------------------------------------------------------------------------ r1091 | sangorrin | 2008-04-08 17:32:45 +0200 (Tue, 08 Apr 2008) | 1 line tab ------------------------------------------------------------------------ r1090 | sangorrin | 2008-04-08 17:32:33 +0200 (Tue, 08 Apr 2008) | 1 line networks are enabled in fna_configuration.h now, not in CFLAGS... i keep the enables in config.mk for the moment but just rules.mk to know which directories should be compiled... with the future omk system i guess this is easier ------------------------------------------------------------------------ r1089 | sangorrin | 2008-04-08 10:45:59 +0200 (Tue, 08 Apr 2008) | 1 line now libfrescan.o instead of .a ------------------------------------------------------------------------ r1088 | sangorrin | 2008-04-08 10:45:24 +0200 (Tue, 08 Apr 2008) | 1 line now libfrescan is a relocatable object file instead of a library .a ------------------------------------------------------------------------ r1087 | sangorrin | 2008-04-08 10:44:54 +0200 (Tue, 08 Apr 2008) | 1 line added fna files, starting its implementation, the rest return an error and write not implemented so far ------------------------------------------------------------------------ r1086 | sangorrin | 2008-04-08 10:44:23 +0200 (Tue, 08 Apr 2008) | 1 line add flag for debugging frescan fna layer ------------------------------------------------------------------------ r1085 | sangorrin | 2008-04-08 10:44:01 +0200 (Tue, 08 Apr 2008) | 1 line add frescan operations ------------------------------------------------------------------------ r1084 | sangorrin | 2008-04-08 10:43:25 +0200 (Tue, 08 Apr 2008) | 1 line add prototypes for frescan ------------------------------------------------------------------------ r1083 | sangorrin | 2008-04-08 10:43:07 +0200 (Tue, 08 Apr 2008) | 1 line now im working on frescan.. ill update rtep if i broke the compilation later ------------------------------------------------------------------------ r1082 | sangorrin | 2008-04-08 10:42:38 +0200 (Tue, 08 Apr 2008) | 1 line now all will be included in libfna.a, the rest will be relocatable linked objects ------------------------------------------------------------------------ r1076 | sangorrin | 2008-04-07 17:58:28 +0200 (Mon, 07 Apr 2008) | 2 lines use frsh.h only, headers mess otherwise ------------------------------------------------------------------------ r1075 | sangorrin | 2008-04-07 17:49:13 +0200 (Mon, 07 Apr 2008) | 1 line include frsh.h to include everything because there were headers problems and miguel has thought it to include only this file ------------------------------------------------------------------------ r1074 | sangorrin | 2008-04-07 17:48:20 +0200 (Mon, 07 Apr 2008) | 1 line frsh_freelist is now part of utils as fadt_freelist ------------------------------------------------------------------------ r1073 | sangorrin | 2008-04-07 17:47:48 +0200 (Mon, 07 Apr 2008) | 1 line utils ------------------------------------------------------------------------ r1068 | sangorrin | 2008-04-07 10:22:18 +0200 (Mon, 07 Apr 2008) | 1 line neg in slave, receive results ------------------------------------------------------------------------ r1067 | sangorrin | 2008-04-07 10:21:55 +0200 (Mon, 07 Apr 2008) | 1 line negotiation reply (TODO: integrate the scheduling test component) ------------------------------------------------------------------------ r1065 | sangorrin | 2008-04-03 18:04:25 +0200 (Thu, 03 Apr 2008) | 1 line negotiation messages, so far a slave call negotiate which sends a negotiate msg which arrives to the master and sends a reply ------------------------------------------------------------------------ r1061 | sangorrin | 2008-04-01 18:09:39 +0200 (Tue, 01 Apr 2008) | 1 line tests for bwres basic ------------------------------------------------------------------------ r1060 | sangorrin | 2008-04-01 18:09:15 +0200 (Tue, 01 Apr 2008) | 1 line negotiator thread and bwres layer ------------------------------------------------------------------------ r1056 | sangorrin | 2008-04-01 11:55:31 +0200 (Tue, 01 Apr 2008) | 1 line makefile for the tests ------------------------------------------------------------------------ r1055 | sangorrin | 2008-04-01 11:55:23 +0200 (Tue, 01 Apr 2008) | 1 line some tests ------------------------------------------------------------------------ r1054 | sangorrin | 2008-04-01 11:55:07 +0200 (Tue, 01 Apr 2008) | 1 line starting to build the negotiations infrastructure ------------------------------------------------------------------------ r1053 | sangorrin | 2008-03-31 13:21:54 +0200 (Mon, 31 Mar 2008) | 1 line frescan compiles ------------------------------------------------------------------------ r1052 | sangorrin | 2008-03-31 13:16:37 +0200 (Mon, 31 Mar 2008) | 1 line starting adaptation to fosa of frescan.. some reordering.. ------------------------------------------------------------------------ r1051 | sangorrin | 2008-03-31 12:24:03 +0200 (Mon, 31 Mar 2008) | 1 line added frescan sources ------------------------------------------------------------------------ r985 | sangorrin | 2008-01-03 12:00:01 +0100 (Thu, 03 Jan 2008) | 1 line changed to N measures ------------------------------------------------------------------------ r984 | sangorrin | 2008-01-03 11:59:36 +0100 (Thu, 03 Jan 2008) | 1 line added send sync operation ------------------------------------------------------------------------ r954 | sangorrin | 2007-12-13 12:58:22 +0100 (Thu, 13 Dec 2007) | 1 line some debugging flags and the target of the makefile ------------------------------------------------------------------------ r944 | sangorrin | 2007-12-11 12:06:25 +0100 (Tue, 11 Dec 2007) | 2 lines new test for measuring the neg time in rtep fna ------------------------------------------------------------------------ r943 | sangorrin | 2007-12-11 12:05:08 +0100 (Tue, 11 Dec 2007) | 1 line makefile changes ------------------------------------------------------------------------ r904 | sangorrin | 2007-11-27 18:53:31 +0100 (Tue, 27 Nov 2007) | 1 line debug flags ------------------------------------------------------------------------ r888 | sangorrin | 2007-11-26 13:06:15 +0100 (Mon, 26 Nov 2007) | 2 lines no qemu image in svn ------------------------------------------------------------------------ r887 | sangorrin | 2007-11-26 13:04:41 +0100 (Mon, 26 Nov 2007) | 1 line backup ------------------------------------------------------------------------ r826 | sangorrin | 2007-10-31 13:28:04 +0100 (Wed, 31 Oct 2007) | 1 line change to unix fna, now we encode streams... but i found a problem in dtm so it wont be useful because multithreaded apps in linux_lib_arch cannot sleep ------------------------------------------------------------------------ r821 | sangorrin | 2007-10-29 19:17:23 +0100 (Mon, 29 Oct 2007) | 1 line debug function ------------------------------------------------------------------------ r820 | sangorrin | 2007-10-29 11:38:31 +0100 (Mon, 29 Oct 2007) | 1 line unix fna, it seems to work ------------------------------------------------------------------------ r819 | sangorrin | 2007-10-26 19:46:58 +0200 (Fri, 26 Oct 2007) | 1 line preparing to do unix fna ------------------------------------------------------------------------ r818 | sangorrin | 2007-10-26 12:23:29 +0200 (Fri, 26 Oct 2007) | 1 line start to do the unix fna ------------------------------------------------------------------------ r817 | sangorrin | 2007-10-26 11:31:44 +0200 (Fri, 26 Oct 2007) | 1 line ------------------------------------------------------------------------ r790 | sangorrin | 2007-10-19 13:44:35 +0200 (Fri, 19 Oct 2007) | 1 line config ------------------------------------------------------------------------ r789 | sangorrin | 2007-10-19 13:43:31 +0200 (Fri, 19 Oct 2007) | 1 line rtep is defined in config.mk ------------------------------------------------------------------------ r788 | sangorrin | 2007-10-19 13:43:15 +0200 (Fri, 19 Oct 2007) | 1 line ifdef rtep ------------------------------------------------------------------------ r784 | sangorrin | 2007-10-19 12:22:48 +0200 (Fri, 19 Oct 2007) | 1 line define rtep ------------------------------------------------------------------------ r783 | sangorrin | 2007-10-19 12:22:41 +0200 (Fri, 19 Oct 2007) | 1 line define rtep ------------------------------------------------------------------------ r772 | sangorrin | 2007-10-19 11:22:16 +0200 (Fri, 19 Oct 2007) | 1 line new makefile ------------------------------------------------------------------------ r771 | sangorrin | 2007-10-19 11:22:00 +0200 (Fri, 19 Oct 2007) | 1 line clean images in tests ------------------------------------------------------------------------ r770 | sangorrin | 2007-10-19 11:20:55 +0200 (Fri, 19 Oct 2007) | 1 line makefile update ------------------------------------------------------------------------ r769 | sangorrin | 2007-10-19 11:20:28 +0200 (Fri, 19 Oct 2007) | 1 line the table is now in fna_configuration.c ------------------------------------------------------------------------ r768 | sangorrin | 2007-10-19 11:20:02 +0200 (Fri, 19 Oct 2007) | 1 line new makefile system ------------------------------------------------------------------------ r767 | sangorrin | 2007-10-19 11:19:13 +0200 (Fri, 19 Oct 2007) | 1 line to store the fna table ------------------------------------------------------------------------ r766 | sangorrin | 2007-10-19 11:18:50 +0200 (Fri, 19 Oct 2007) | 1 line no elif ------------------------------------------------------------------------ r765 | sangorrin | 2007-10-19 11:18:30 +0200 (Fri, 19 Oct 2007) | 1 line define rtep ------------------------------------------------------------------------ r764 | sangorrin | 2007-10-19 11:15:00 +0200 (Fri, 19 Oct 2007) | 1 line que no ------------------------------------------------------------------------ r733 | telleriam | 2007-09-28 13:44:34 +0200 (Fri, 28 Sep 2007) | 3 lines Global update in FNA to update license text ------------------------------------------------------------------------ r698 | sangorrin | 2007-09-06 17:15:20 +0200 (Thu, 06 Sep 2007) | 1 line no messages ------------------------------------------------------------------------ r689 | sangorrin | 2007-09-03 19:38:38 +0200 (Mon, 03 Sep 2007) | 1 line makefile with frsh ------------------------------------------------------------------------ r688 | sangorrin | 2007-09-03 19:38:11 +0200 (Mon, 03 Sep 2007) | 1 line send receive ------------------------------------------------------------------------ r687 | sangorrin | 2007-09-03 19:37:21 +0200 (Mon, 03 Sep 2007) | 1 line check fna with frsh running ------------------------------------------------------------------------ r686 | sangorrin | 2007-09-03 19:37:07 +0200 (Mon, 03 Sep 2007) | 1 line comment until implemented ------------------------------------------------------------------------ r685 | sangorrin | 2007-09-03 19:36:07 +0200 (Mon, 03 Sep 2007) | 1 line chan ------------------------------------------------------------------------ r668 | sangorrin | 2007-08-10 19:05:44 +0200 (Fri, 10 Aug 2007) | 1 line renegotiate_sync ------------------------------------------------------------------------ r645 | sangorrin | 2007-08-09 13:47:44 +0200 (Thu, 09 Aug 2007) | 1 line big commint with a lot of changes, see files ------------------------------------------------------------------------ r644 | sangorrin | 2007-08-09 13:46:06 +0200 (Thu, 09 Aug 2007) | 1 line lib for rtepfna ------------------------------------------------------------------------ r643 | sangorrin | 2007-08-09 13:40:01 +0200 (Thu, 09 Aug 2007) | 1 line hooks for fna ------------------------------------------------------------------------ r635 | sangorrin | 2007-08-02 12:23:10 +0200 (Thu, 02 Aug 2007) | 1 line assert ------------------------------------------------------------------------ r510 | sangorrin | 2007-06-25 17:34:48 +0200 (Mon, 25 Jun 2007) | 2 lines fna tests ------------------------------------------------------------------------ r509 | sangorrin | 2007-06-25 17:32:52 +0200 (Mon, 25 Jun 2007) | 1 line ------------------------------------------------------------------------ r501 | sangorrin | 2007-06-19 09:46:07 +0200 (Tue, 19 Jun 2007) | 1 line i add FROM when receiving ------------------------------------------------------------------------ r407 | sangorrin | 2007-05-15 19:40:41 +0200 (Tue, 15 May 2007) | 2 lines update doc ------------------------------------------------------------------------ r362 | sangorrin | 2007-03-22 11:33:07 +0100 (Thu, 22 Mar 2007) | 1 line create_callback ------------------------------------------------------------------------ r357 | sangorrin | 2007-03-20 18:27:20 +0100 (Tue, 20 Mar 2007) | 2 lines maxsize ------------------------------------------------------------------------ r348 | sangorrin | 2007-03-20 11:27:55 +0100 (Tue, 20 Mar 2007) | 2 lines doc update ------------------------------------------------------------------------ r347 | sangorrin | 2007-03-20 11:26:04 +0100 (Tue, 20 Mar 2007) | 1 line correct network functions ------------------------------------------------------------------------ r344 | sangorrin | 2007-03-19 17:16:50 +0100 (Mon, 19 Mar 2007) | 1 line functions to map bytes to budgets changed and a new function for effective budget ------------------------------------------------------------------------ r334 | sangorrin | 2007-03-14 19:03:04 +0100 (Wed, 14 Mar 2007) | 1 line Changes in names of spare capacity functions. Change in message_get_tx_time (tx_time is the physical time without overheads) ------------------------------------------------------------------------ r332 | sangorrin | 2007-03-13 19:16:36 +0100 (Tue, 13 Mar 2007) | 2 lines update doc ------------------------------------------------------------------------ r331 | sangorrin | 2007-03-13 12:20:48 +0100 (Tue, 13 Mar 2007) | 1 line changes in get_status functions and add receive create endpoint ------------------------------------------------------------------------ r321 | sangorrin | 2007-03-09 10:11:06 +0100 (Fri, 09 Mar 2007) | 2 lines copying, readme, todo.. ------------------------------------------------------------------------ r320 | sangorrin | 2007-03-09 10:04:15 +0100 (Fri, 09 Mar 2007) | 2 lines src is source common for all protocols. each protocol has its source in its own dir ------------------------------------------------------------------------ r319 | sangorrin | 2007-03-09 09:53:45 +0100 (Fri, 09 Mar 2007) | 2 lines change source names because they don't contain the whole code of the protocols ------------------------------------------------------------------------ r318 | sangorrin | 2007-03-07 10:30:54 +0100 (Wed, 07 Mar 2007) | 2 lines first upload of fna ------------------------------------------------------------------------ r317 | sangorrin | 2007-03-07 10:19:38 +0100 (Wed, 07 Mar 2007) | 2 lines creating trunk ------------------------------------------------------------------------