)]}' { "log": [ { "commit": "c9eaf17341834de00351bf79f16b2d879c8aea96", "tree": "d8b2005197444fa6b6bdf8e8c8fd6eaf2db9ecd7", "parents": [ "4ec93edb14fe5fdee9fae6335f2cbba204627eac" ], "author": { "name": "YOSHIFUJI Hideaki", "email": "yoshfuji@linux-ipv6.org", "time": "Fri Feb 09 23:24:38 2007 +0900" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Feb 10 23:19:27 2007 -0800" }, "message": "[NET] DCCP: Fix whitespace errors.\n\nSigned-off-by: YOSHIFUJI Hideaki \u003cyoshfuji@linux-ipv6.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0f08461ebf89e10f7db9042fb028359b810c3c81", "tree": "3a8bdeebfd307483d3a6ab094a770449112b4135", "parents": [ "919afbd68863550665b328a78107bc2919c5e3f4" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Mon Feb 05 18:18:21 2007 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Thu Feb 08 12:38:56 2007 -0800" }, "message": "[DCCP]: Warning fixes.\n\nnet/dccp/ccids/ccid3.c: In function `ccid3_hc_rx_packet_recv\u0027:\nnet/dccp/ccids/ccid3.c:1007: warning: long int format, different type arg (arg 3)\nnet/dccp/ccids/ccid3.c:1007: warning: long int format, different type arg (arg 4)\n\nopaque types must be suitably cast for printing.\n\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "832e3ca62d6ee1893db05f5202c24ae5331e19ba", "tree": "c6cf0507bd9a4e48363748b0a2f2c89ad40ab6ab", "parents": [ "e1b4b9f3986b80d5785d91dbd8d72cfaf9fd1117" ], "author": { "name": "Ian McDonald", "email": "ian.mcdonald@jandi.co.nz", "time": "Tue Dec 12 00:47:59 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Wed Dec 13 16:48:24 2006 -0800" }, "message": "[DCCP] ccid3: return value in ccid3_hc_rx_calc_first_li\n\nIn a recent patch we introduced invalid return codes which will result in the\nopposite of what is intended (i.e. send more packets in face of peculiar\nnetwork conditions).\n\nThis fixes it by returning ~0 which means not calculated as per\ndccp_li_hist_calc_i_mean.\n\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "8109b02b5397ed52a32c116163a62a34f4768b26", "tree": "0dbeae8cc999d2f6b71807b942eda7ff686ba7d2", "parents": [ "1fba78b6cba14bd37fdb12c5367f1e4d58ff2e0f" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 10 16:01:18 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:35:00 2006 -0800" }, "message": "[DCCP]: Whitespace cleanups\n\nThat accumulated over the last months hackaton, shame on me for not\nusing git-apply whitespace helping hand, will do that from now on.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "1fba78b6cba14bd37fdb12c5367f1e4d58ff2e0f", "tree": "dc606497372ac469f698fa2a31d956a67a62d739", "parents": [ "9e8efc824098c241a1cde81f5558d222cb6f9369" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 10 15:39:29 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:59 2006 -0800" }, "message": "[DCCP] ccid3: Fixup some type conversions related to rtts\n\nSpotted by David Miller when compiling on sparc64, I reproduced it here on\nparisc64, that are the only platforms to define __kernel_suseconds_t as an\n\u0027int\u0027, all the others, x86_64 and x86 included typedef it as a \u0027long\u0027, but from\nthe definition of suseconds_t it should just be an \u0027int\u0027 on platforms where it\nis \u003e\u003d 32bits, it would not require all the castings from suseconds_t to (int)\nwhen printking variables of this type, that are not needed on parisc64 and\nsparc64.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "9e8efc824098c241a1cde81f5558d222cb6f9369", "tree": "443c107edcac509a93857184f1d2e00933369dcb", "parents": [ "7af5af3013f84693a7af581e3c8b32f9db493cf5" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:24:57 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:58 2006 -0800" }, "message": "[DCCP] ccid3: BUG-FIX - conversion errors\n\nThis fixes conversion errors which arose by not properly type-casting\nfrom u32 to __u64. Fixed by explicitly casting each type which is not\n__u64, or by performing operation after assignment.\n\nThe patch further adds missing debug information to track the current\nvalue of X_recv.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "7af5af3013f84693a7af581e3c8b32f9db493cf5", "tree": "4434c3993e0a6da3cd5b5ade8db06c9491773ade", "parents": [ "85dcb1f78039cc55ccddd9622657a338b718aafe" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:24:33 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:57 2006 -0800" }, "message": "[DCCP] ccid3: Reorder packet history source file\n\nNo code change at all.\n\nThis reorders the source file to follow the same order as the corresponding\nheader file.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "85dcb1f78039cc55ccddd9622657a338b718aafe", "tree": "66cf39fbb6c147a013588142ba5488fcd8e666d9", "parents": [ "a967241129f6dc1db92fee2c808f73a90f5f47a3" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:24:11 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:56 2006 -0800" }, "message": "[DCCP] ccid3: Reorder packet history header file\n\nNo code change at all.\n\nTo make the header file easier to read, the following ordering is established\namong the declarations:\n\t* hist_new\n\t* hist_delete\n\t* hist_entry_new\n\t* hist_head\n\t* hist_find_entry\n\t* hist_add_entry\n\t* hist_entry_delete\n\t* hist_purge\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "a967241129f6dc1db92fee2c808f73a90f5f47a3", "tree": "7fb9aaedfcd47260b341050bb006729df5c6ec87", "parents": [ "c5a1ae9a4cc4aef1505fa0aba079f834c3ee1af1" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:14:12 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:55 2006 -0800" }, "message": "[DCCP] ccid3: Make debug output consistent\n\nThis patch does not alter any algorithm, just the debug message format:\n\n * s#%s, sk\u003d%p#%s(%p)#g\n\n * when a statename is present, it now uses %s(%p, state\u003d%s)\n\n * when only function entry is debugged, it adds an `- entry\u0027\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "c5a1ae9a4cc4aef1505fa0aba079f834c3ee1af1", "tree": "ce26c29076ae01f762fd5b09becca79ac602a63b", "parents": [ "e312d100f19fdfe1019512b07a9d15653f254abf" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:09:21 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:54 2006 -0800" }, "message": "[DCCP] ccid3: Perform history operations only after packet has been sent\n\n This migrates all packet history operations into the routine\n ccid3_hc_tx_packet_sent, thereby removing synchronization problems\n that occur when, as before, the operations are spread over multiple\n routines.\n The following minor simplifications are also applied:\n * several simplifications now follow from this change - several tests\n are now no longer required\n * removal of one unnecessary variable (dp)\n\nJustification:\n\n Currently packet history operations span two different routines,\n one of which is likely to pass through several iterations of sleeping\n and awakening.\n The first routine, ccid3_hc_tx_send_packet, allocates an entry and\n sets a few fields. The remaining fields are filled in when the second\n routine (which is not within a sleeping context), ccid3_hc_tx_packet_sent,\n is called. This has several strong drawbacks:\n * it is not necessary to split history operations - all fields can be\n filled in by the second routine\n * the first routine is called multiple times, until a packet can be sent,\n and sleeps meanwhile - this causes a lot of difficulties with regard to\n keeping the list consistent\n * since both routines do not have a producer-consumer like synchronization,\n it is very difficult to maintain data across calls to these routines\n * the fact that the routines are called in different contexts (sleeping, not\n sleeping) adds further problems\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "e312d100f19fdfe1019512b07a9d15653f254abf", "tree": "96b3275b746b2c871ab3d2d0f6e1d683ea0ae030", "parents": [ "9f8681db961de6d75e43844e9664f1f81e05e1b7" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:08:09 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:53 2006 -0800" }, "message": "[DCCP] ccid3: TX history - remove unused field\n\nThis removes the `dccphtx_ccval\u0027 field since it is nowhere used in the code and\nin fact not necessary for the accounting.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "9f8681db961de6d75e43844e9664f1f81e05e1b7", "tree": "07679d0ea408127119bc8b2ce74b38031da204e5", "parents": [ "de553c189e3faa0d0c38f366f73379b46587b80e" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:07:37 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:52 2006 -0800" }, "message": "[DCCP] ccid3: Shift window counter computation\n\n This puts the window counter computation [RFC 4342, 8.1] into a separate\n function which is called whenever a new packet is ready for immediate\n transmission in ccid3_hc_tx_send_packet.\n\nJustification:\n\n The window counter update was previously computed after the packet was sent. This has\n two drawbacks, both fixed by this patch:\n 1) re-compute another timestamp almost directly after the packet was sent (expensive),\n 2) the CCVal for the window counter is needed at the instant the packet is sent.\n\nFurther details:\n\n The initialisation of the window counter is left in the state NO_SENT, as before.\n The algorithm will do nothing if either RTT is initialised to 0 (which is ok) or if\n the RTT value remains below 4 microseconds (which is almost pathological).\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "de553c189e3faa0d0c38f366f73379b46587b80e", "tree": "0fbc7bd2c042518383ee07287812453176ae8b0d", "parents": [ "fe0499ae95f5f636bda1f6e0bdba5b7b023ea827" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:06:32 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:51 2006 -0800" }, "message": "[DCCP] ccid3: Sanity-check RTT samples\n\nCCID3 performance depends much on the accuracy of RTT samples. If RTT\nsamples grow too large, performance can be catastrophically poor.\n\nTo limit the amount of possible damage in such cases, the patch\n * introduces an upper limit which identifies a maximum `sane\u0027 RTT value;\n * uses a macro to enforce this upper limit.\n\nUsing a macro was given preference, since it is necessary to identify the\ncalling function in the warning message. Since exceeding this threshold\nidentifies a critical condition, DCCP_CRIT is used and not DCCP_WARN.\n\nMany thanks to Ian McDonald for collaboration on this issue.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "fe0499ae95f5f636bda1f6e0bdba5b7b023ea827", "tree": "d944ebcdaa0b689de58e3c98adf3d3b707385493", "parents": [ "65d6c2b42ecc0ca77bb3c0fe27fcbdc84fc7a79a" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:06:01 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:50 2006 -0800" }, "message": "[DCCP] ccid3: Initialise RTT values\n\nIn both the sender and the receiver it is possible that the stored\nRTT value is accessed before an actual RTT estimate has been computed.\n\nThis patch\n * initialises the sender RTT to 0\n - the sender always accesses the RTT in ccid3_hc_tx_packet_sent\n - the RTT is further needed for the window counter algorithm\n\n * replaces the receiver initialisation of 5msec with 0\n - which has the same effect and removes an `XXX\u0027\n - the RTT value is needed in ccid3_hc_rx_packet_recv as rtt_prev\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "65d6c2b42ecc0ca77bb3c0fe27fcbdc84fc7a79a", "tree": "f38a54545125cf0c736fd484f03ec3e3ee3d971a", "parents": [ "f6282f4da5eb67e08eff94d73a1911c3a008846a" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:05:38 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:49 2006 -0800" }, "message": "[DCCP] ccid: Deprecate ccid_hc_tx_insert_options\n\n The function ccid3_hc_tx_insert_options only does a redundant no-op,\n as the operation\n\n DCCP_SKB_CB(skb)-\u003edccpd_ccval \u003d hctx-\u003eccid3hctx_last_win_count;\n\n is already performed _unconditionally_ in ccid3_hc_tx_send_packet.\n\n Since there is further no current need for this function, it is removed\n entirely. Since furthermore, there is actually no present need for the\n entire interface function ccid_hc_tx_insert_options, it was decided to\n remove it also, to clean up the interface.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "bf58a381e8106fe73247c753e3da58fcb5eabd2e", "tree": "4f0539a3d96930bddc5291b136f4fa08c1af8e77", "parents": [ "d63d8364cfe17fc9bb738f554f452595f76f21d2" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:04:43 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:47 2006 -0800" }, "message": "[DCCP]: Only deliver to the CCID rx side in charge\n\nThis is an optimisation to reduce CPU load. The received feedback is now\nonly directed to the active CCID component, without requiring processing\nalso by the inactive one.\n\nAs a consequence, a similar test in ccid3.c is now redundant and is\nalso removed.\n\nJustification:\n\n Currently DCCP works as a unidirectional service, i.e. a listening server\n is not at the same time a connecting client.\n As far as I can see, several modifications are necessary until that\n becomes possible.\n At the present time, received feedback is both fed to the rx/tx CCID\n modules. In unidirectional service, only one of these is active at any\n one time.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "d63d8364cfe17fc9bb738f554f452595f76f21d2", "tree": "32bd2371deb9ddc8a7c7bd835f33c2290cc0e8bf", "parents": [ "0f9e5b573f7249b0e04a03457b55081d1f60f2bf" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:04:16 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:46 2006 -0800" }, "message": "[DCCP]: Simplify TFRC calculation\n\nIn migrating towards using the newer functions scaled_div/scaled_div32\nfor TFRC computations mapped from floating-point onto integer arithmetic,\nthis completes the last stage of modifications.\n\nIn particular, the overflow case for computing X_calc is circumvented by\n * breaking the computation into two stages\n * the first stage, res \u003d (s*1E6)/R, cannot overflow due to use of u64\n * in the second stage, res \u003d (res*1E6)/f, overflow on u32 is avoided due\n to (i) returning UINT_MAX in this case (which is logically appropriate)\n and (ii) issuing a warning message into the system log (since very likely\n there is a problem somewhere else with the parameters)\n\nLastly, all such scaling operations are now exported into tfrc.h, since\nactually this form of scaled computation is specific to TFRC and not to CCID3.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "0f9e5b573f7249b0e04a03457b55081d1f60f2bf", "tree": "fa33e3252d5df520348bb22d2a14aeac0e6b2bfc", "parents": [ "bfe24a6cc222d27e1491f850802fa6932232b8ef" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:03:51 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:45 2006 -0800" }, "message": "[DCCP]: Debug timeval operations\n\nProblem:\n\n Most target types in the CCID3 code are u32, so subtle conversion errors\n can occur if signed time calculations yield negative results: the original\n values are lost in the conversion to unsigned, calculation errors go undetected.\n\n This patch therefore\n * sets all critical time types from unsigned to suseconds_t\n * avoids comparison between signed/unsigned via type-casting\n * provides ample warning messages in case time calculations are negative\n\n These warning messages can be removed at a later stage when the code\n has undergone more testing.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "bfe24a6cc222d27e1491f850802fa6932232b8ef", "tree": "a1782cdb20b19ffb07e1e7ce1ab7aba34fddcecd", "parents": [ "b9039a2a8df974d7702564318722434bb276a995" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:03:30 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:44 2006 -0800" }, "message": "[DCCP] ccid3: Simplify calculation for reverse lookup of p\n\n This simplifies the calculation of a value p for a given fval when the\n first loss interval is computed (RFC 3448, 6.3.1). It makes use of the\n two new functions scaled_div/scaled_div32 to provide overflow protection.\n\n Additionally, protection against divide-by-zero is extended - in this\n case the function will return the maximally possible value of p\u003d100%.\n\nBackground:\n\n The maximum fval, f(100%), is approximately 244, i.e. the scaled value of fval\n should never exceed 244E6, which fits easily into u32. The problem is the scaling\n by 10^6, since additionally R(TT) is in microseconds.\n This is resolved by breaking the division into two stages: the first stage\n computes fval\u003d(s*10^6)/R, stores that into u64; the second stage computes\n fval \u003d (fval*10^6)/X_recv and complains if overflow is reached for u32.\n This case is safe since the TFRC reverse-lookup routine then returns p\u003d100%.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "b9039a2a8df974d7702564318722434bb276a995", "tree": "ee67d87ec5b6e0a7f3100f63c31050270593146d", "parents": [ "1a21e49a8d60f588c1276f765198b14d5688a778" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:02:51 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:43 2006 -0800" }, "message": "[DCCP] ccid3: Replace scaled division operations\n\nThis replaces the remaining uses of usecs_div with scaled_div32, which\ninternally uses 64bit division and produces a warning on overflow.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "1a21e49a8d60f588c1276f765198b14d5688a778", "tree": "4b8770be71047437f30da694649719948977b189", "parents": [ "179ebc9f92da88e15ea86d7d27308c92712d8ee9" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:02:12 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:42 2006 -0800" }, "message": "[DCCP] ccid3: Finer-grained resolution of sending rates\n\nThis patch\n * resolves a bug where packets smaller than 32/64 bytes resulted in sending rates of 0\n * supports all sending rates from 1/64 bytes/second up to 4Gbyte/second\n * simplifies the present overflow problems in calculations\n\nCurrent sending rate X and the cached value X_recv of the receiver-estimated\nsending rate are both scaled by 64 (2^6) in order to\n * cope with low sending rates (minimally 1 byte/second)\n * allow upgrading to use a packets-per-second implementation of CCID 3\n * avoid calculation errors due to integer arithmetic cut-off\n\nThe patch implements a revised strategy from\nhttp://www.mail-archive.com/dccp@vger.kernel.org/msg01040.html\n\nThe only difference with regard to that strategy is that t_ipi is already\nused in the calculation of the nofeedback timeout, which saves one division.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "179ebc9f92da88e15ea86d7d27308c92712d8ee9", "tree": "6b8a18fd9e5b3447b8b32560d258f92276c358b7", "parents": [ "ff58629824c068e2a75402b5b83f12af0b36d394" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:01:22 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:41 2006 -0800" }, "message": "[DCCP] ccid3: Fix two bugs in sending rate computation\n\nThis fixes\n 1) a bug in the recomputation of the sending rate by the nofeedback\n timer when no feedback at all has so far been sent by the receiver:\n min_t was used instead of max_t, which is wrong (cf. RFC 3448, p. 10);\n\n 2) an error in the computation of larger initial windows: instead of\n min(... max()) (cf. RFC 4342, 5.), the code had used max(... max()).\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "ff58629824c068e2a75402b5b83f12af0b36d394", "tree": "0d8f4a64dbe4e52879cac33174df406adf77da44", "parents": [ "45393a66a2c30fc8ed86e6c79a63a3cd5231c69f" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 10 00:00:14 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:40 2006 -0800" }, "message": "[DCCP] ccid3: Two optimisations for sending rate recomputation\n\nThis performs two optimisations for the recomputation of the sending rate.\n\n1) Currently the target sending rate X_calc is recalculated whenever\n\ta) the nofeedback timer expires, or\n\tb) a feedback packet is received.\n In the (a) case, recomputing X_calc is redundant, since\n\n * the parameters p and RTT do not change in between the\n reception of feedback packets;\n\n * the parameter X_recv is either modified from received\n feedback or via the nofeedback timer;\n\n * a test (`p \u003d\u003d 0\u0027) in the nofeedback timer avoids using\n a stale/undefined value of X_calc if p was previously 0.\n\n2) The nofeedback timer now only recomputes a timestamp when p \u003d\u003d 0.\n This is according to step (4) of [RFC 3448, 4.3] and avoids\n unnecessarily determining a timestamp.\n\nA debug statement about not updating X is also removed - it helps very\nlittle in debugging and just clutters the logs.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "45393a66a2c30fc8ed86e6c79a63a3cd5231c69f", "tree": "628feedd3d6d1fbdb1728d57e4bb62e1fc61e5c4", "parents": [ "5fce9a2da8a3b4ed088d9c0ac7c938a638044fdb" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sat Dec 09 23:59:14 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Dec 11 14:34:39 2006 -0800" }, "message": "[DCCP] ccid3: Check against too large p\n\nThis patch follows a suggestion by Ian McDonald and ensures that in\nthe current code the value of p can not exceed 100%. Such a value is\nillegal and would consequently cause a bug condition in tfrc_calc_x().\n\nThe receiver case is also tested, and a warning message is added.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "e18b890bb0881bbab6f4f1a6cd20d9c60d66b003", "tree": "4828be07e1c24781c264b42c5a75bcd968223c3f", "parents": [ "441e143e95f5aa1e04026cb0aa71c801ba53982f" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Wed Dec 06 20:33:20 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.osdl.org", "time": "Thu Dec 07 08:39:25 2006 -0800" }, "message": "[PATCH] slab: remove kmem_cache_t\n\nReplace all uses of kmem_cache_t with struct kmem_cache.\n\nThe patch was generated using the following script:\n\n\t#!/bin/sh\n\t#\n\t# Replace one string by another in all the kernel sources.\n\t#\n\n\tset -e\n\n\tfor file in `find * -name \"*.c\" -o -name \"*.h\"|xargs grep -l $1`; do\n\t\tquilt add $file\n\t\tsed -e \"1,\\$s/$1/$2/g\" $file \u003e/tmp/$$\n\t\tmv /tmp/$$ $file\n\t\tquilt refresh\n\tdone\n\nThe script was run like this\n\n\tsh replace kmem_cache_t \"struct kmem_cache\"\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "54e6ecb23951b195d02433a741c7f7cb0b796c78", "tree": "c8885c49f37c8d383945b8af69d51597494ed62c", "parents": [ "f7267c0c0721fd02ad3dc37c3d6dd24ccd81d4d6" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Wed Dec 06 20:33:16 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.osdl.org", "time": "Thu Dec 07 08:39:24 2006 -0800" }, "message": "[PATCH] slab: remove SLAB_ATOMIC\n\nSLAB_ATOMIC is an alias of GFP_ATOMIC\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "2bbf29acd8f7adcf161de7e5d891b4095687a59f", "tree": "955b15fa3bcdee1f879949d8639c5e9d5b82e84e", "parents": [ "44158306d756c88272c8faf243ca68897498e219" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:53:27 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:53:27 2006 -0200" }, "message": "[DCCP] tfrc: Binary search for reverse TFRC lookup\n\nThis replaces the linear search algorithm for reverse lookup with\nbinary search.\n\nIt has the advantage of better scalability: O(log2(N)) instead of O(N).\nThis means that the average number of iterations is reduced from 250\n(linear search if each value appears equally likely) down to at most 9.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "44158306d756c88272c8faf243ca68897498e219", "tree": "b5406da158950a099336a67c507a0dd55c04fa25", "parents": [ "006042d7e1a0aae35c9dd9eb8ec71fa379679adb" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:53:07 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:53:07 2006 -0200" }, "message": "[DCCP] ccid3: Deprecate TFRC_SMALLEST_P\n\n This patch deprecates the existing use of an arbitrary value TFRC_SMALLEST_P\n for low-threshold values of p. This avoids masking low-resolution errors.\n Instead, the code now checks against real boundaries (implemented by preceding\n patch) and provides warnings whenever a real value falls below the threshold.\n\n If such messages are observed, it is a better solution to take this as an\n indication that the lookup table needs to be re-engineered.\n\nChangelog:\n----------\n This patch\n * makes handling all TFRC resolution errors local to the TFRC library\n\n * removes unnecessary test whether X_calc is \u0027infinity\u0027 due to p\u003d\u003d0 -- this\n condition is already caught by tfrc_calc_x()\n\n * removes setting ccid3hctx_p \u003d TFRC_SMALLEST_P in ccid3_hc_tx_packet_recv\n since this is now done by the TFRC library\n\n * updates BUG_ON test in ccid3_hc_tx_no_feedback_timer to take into account\n that p now is either 0 (and then X_calc is irrelevant), or it is \u003e 0; since\n the handling of TFRC_SMALLEST_P is now taken care of in the tfrc library\n\nJustification:\n--------------\n The TFRC code uses a lookup table which has a bounded resolution.\n The lowest possible value of the loss event rate `p\u0027 which can be\n resolved is currently 0.0001. Substituting this lower threshold for\n p when p is less than 0.0001 results in a huge, exponentially-growing\n error. The error can be computed by the following formula:\n\n (f(0.0001) - f(p))/f(p) * 100 for p \u003c 0.0001\n\n Currently the solution is to use an (arbitrary) value\n TFRC_SMALLEST_P \u003d 40 * 1E-6 \u003d 0.00004\n and to consider all values below this value as `virtually zero\u0027. Due to\n the exponentially growing resolution error, this is not a good idea, since\n it hides the fact that the table can not resolve practically occurring cases.\n Already at p \u003d\u003d TFRC_SMALLEST_P, the error is as high as 58.19%!\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "006042d7e1a0aae35c9dd9eb8ec71fa379679adb", "tree": "e0f7a4f351fbf147bd452ff45061245fd985f27d", "parents": [ "8d0086adac0041de66b5f41b77eec0d8d239e16c" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:52:41 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:52:41 2006 -0200" }, "message": "[DCCP] tfrc: Identify TFRC table limits and simplify code\n\nThis\n * adds documentation about the lowest resolution that is possible within\n the bounds of the current lookup table\n * defines a constant TFRC_SMALLEST_P which defines this resolution\n * issues a warning if a given value of p is below resolution\n * combines two previously adjacent if-blocks of nearly identical\n structure into one\n\nThis patch does not change the algorithm as such.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "8d0086adac0041de66b5f41b77eec0d8d239e16c", "tree": "9830f9188a9ba57bf75c0fa7c2246866f8a8789a", "parents": [ "90fb0e60dd9178dbca2e42c682c483cdb7ea9f2d" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:52:26 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:52:26 2006 -0200" }, "message": "[DCCP] tfrc: Add protection against invalid parameters to TFRC routines\n\n 1) For the forward X_calc lookup, it\n * protects effectively against RTT\u003d0 (this case is possible), by\n returning the maximal lookup value instead of just setting it to 1\n * reformulates the array-bounds exceeded condition: this only happens\n if p is greater than 1E6 (due to the scaling)\n * the case of negative indices can now with certainty be excluded,\n since documentation shows that the formulas are within bounds\n * additional protection against p \u003d 0 (would give divide-by-zero)\n\n 2) For the reverse lookup, it warns against\n * protects against exceeding array bounds\n * now returns 0 if f(p) \u003d 0, due to function definition\n * warns about minimal resolution error and returns the smallest table\n value instead of p\u003d0 [this would mask congestion conditions]\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "90fb0e60dd9178dbca2e42c682c483cdb7ea9f2d", "tree": "09e750ded7fb3b34a0b146ef96a03a9495a9c37f", "parents": [ "50ab46c790a3408c441ba4c2faa9555cacc20028" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:52:01 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:52:01 2006 -0200" }, "message": "[DCCP] tfrc: Fix small error in reverse lookup of p for given f(p)\n\nThis fixes the following small error in tfrc_calc_x_reverse_lookup.\n\n 1) The table is generated by the following equations:\n\tlookup[index][0] \u003d g((index+1) * 1000000/TFRC_CALC_X_ARRSIZE);\n\tlookup[index][1] \u003d g((index+1) * TFRC_CALC_X_SPLIT/TFRC_CALC_X_ARRSIZE);\n where g(q) is 1E6 * f(q/1E6)\n\n 2) The reverse lookup assigns an entry in lookup[index][small]\n\n 3) This index needs to match the above, i.e.\n * if small\u003d0 then\n\n \t\tp \u003d (index + 1) * 1000000/TFRC_CALC_X_ARRSIZE\n\n * if small\u003d1 then\n\n\t\tp \u003d (index+1) * TFRC_CALC_X_SPLIT/TFRC_CALC_X_ARRSIZE\n\nThese are exactly the changes that the patch makes; previously the code did\nnot conform to the way the lookup table was generated (this difference resulted\nin a mean error of about 1.12%).\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "50ab46c790a3408c441ba4c2faa9555cacc20028", "tree": "c1d1a93375908ad66fb6599532351bfcb3785269", "parents": [ "26af3072b035daadf34a99d02510f0ff98f41f90" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:51:29 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:51:29 2006 -0200" }, "message": "[DCCP] tfrc: Document boundaries and limits of the TFRC lookup table\n\nThis adds documentation for the TCP Reno throughput equation which is at\nthe heart of the TFRC sending rate / loss rate calculations.\n\nIt spells out precisely how the values were determined and what they mean.\nThe equations were derived through reverse engineering and found to be\nfully accurate (verified using test programs).\n\nThis patch does not change any code.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "26af3072b035daadf34a99d02510f0ff98f41f90", "tree": "2e3bde9b1b8c4bf9f9f74f0b7c2ee0f512664ecf", "parents": [ "5c3fbb6acf9d32772ec7fc01cedd9478d0e26f44" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:51:14 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:51:14 2006 -0200" }, "message": "[DCCP] ccid3: Fix warning message about illegal ACK\n\nThis avoids a (harmless) warning message being printed at the DCCP server\n(the receiver of a DCCP half connection).\n\nIncoming packets are both directed to\n\n * ccid_hc_rx_packet_recv() for the server half\n * ccid_hc_tx_packet_recv() for the client half\n\nThe message gets printed since on a server the client half is currently not\nsending data packets.\nThis is resolved for the moment by checking the DCCP-role first. In future\ntimes (bidirectional DCCP connections), this test may have to be more\nsophisticated.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "5c3fbb6acf9d32772ec7fc01cedd9478d0e26f44", "tree": "25d83ea9d776714b15d172aebb861e5d3f49f31a", "parents": [ "76d127779e51fbf67ad6793214369aa1fea90278" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:50:56 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:50:56 2006 -0200" }, "message": "[DCCP] ccid3: Fix bug in calculation of send rate\n\nThe main object of this patch is the following bug:\n \u003d\u003d\u003e In ccid3_hc_tx_packet_recv, the parameters p and X_recv were updated\n _after_ the send rate was calculated. This is clearly an error and is\n resolved by re-ordering statements.\n\nIn addition,\n * r_sample is converted from u32 to long to check whether the time difference\n was negative (it would otherwise be converted to a large u32 value)\n * protection against RTT\u003d0 (this is possible) is provided in a further patch\n * t_elapsed is also converted to long, to match the type of r_sample\n * adds a a more debugging information regarding current send rates\n * various trivial comment/documentation updates\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "76d127779e51fbf67ad6793214369aa1fea90278", "tree": "cc6cbeb3b7eb6c59b5e13bb12a74e0fc89988c78", "parents": [ "8a508ac26e72aa203677aa6a8464bd3ea44216a6" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:50:42 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:50:42 2006 -0200" }, "message": "[DCCP]: Fix BUG in retransmission delay calculation\n\nThis bug resulted in ccid3_hc_tx_send_packet returning negative\ndelay values, which in turn triggered silently dequeueing packets in\ndccp_write_xmit. As a result, only a few out of the submitted packets made\nit at all onto the network. Occasionally, when dccp_wait_for_ccid was\ninvolved, this also triggered a bug warning since ccid3_hc_tx_send_packet\nreturned a negative value (which in reality was a negative delay value).\n\nThe cause for this bug lies in the comparison\n\n if (delay \u003e\u003d hctx-\u003eccid3hctx_delta)\n\treturn delay / 1000L;\n\nThe type of `delay\u0027 is `long\u0027, that of ccid3hctx_delta is `u32\u0027. When comparing\nnegative long values against u32 values, the test returned `true\u0027 whenever delay\nwas smaller than 0 (meaning the packet was overdue to send).\n\nThe fix is by casting, subtracting, and then testing the difference with\nregard to 0.\n\nThis has been tested and shown to work.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "8a508ac26e72aa203677aa6a8464bd3ea44216a6", "tree": "5a2167652960e2dc7300ad6ff3247bc0c31eed5c", "parents": [ "2b5f6dcce5bf94b9b119e9ed8d537098ec61c3d2" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Sun Dec 03 14:50:23 2006 -0200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Dec 03 14:50:23 2006 -0200" }, "message": "[DCCP]: Use higher RTO default for CCID3\n\nThe TFRC nofeedback timer normally expires after the maximum of 4\nRTTs and twice the current send interval (RFC 3448, 4.3). On LANs\nwith a small RTT this can mean a high processing load and reduced\nperformance, since then the nofeedback timer is triggered very\nfrequently.\n\nThis patch provides a configuration option to set the bound for the\nnofeedback timer, using as default 100 milliseconds.\n\nBy setting the configuration option to 0, strict RFC 3448 behaviour\ncan be enforced for the nofeedback timer.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "6b57c93dc3aa0115b589cb89ef862d46ab9bd95e", "tree": "f825b1288d7839ade26225bb0a630540e38ffe41", "parents": [ "a79ef76f4d8424324c2f108824a7398571193f43" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Tue Nov 28 19:55:06 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:31:02 2006 -0800" }, "message": "[DCCP]: Use `unsigned\u0027 for packet lengths\n\nThis patch implements a suggestion by Ian McDonald and\n\n 1) Avoids tests against negative packet lengths by using unsigned int\n for packet payload lengths in the CCID send_packet()/packet_sent() routines\n\n 2) As a consequence, it removes an now unnecessary test with regard to `len \u003e 0\u0027\n in ccid3_hc_tx_packet_sent: that condition is always true, since\n * negative packet lengths are avoided\n * ccid3_hc_tx_send_packet flags an error whenever the payload length is 0.\n As a consequence, ccid3_hc_tx_packet_sent is never called as all errors\n returned by ccid_hc_tx_send_packet are caught in dccp_write_xmit\n\n 3) Removes the third argument of ccid_hc_tx_send_packet (the `len\u0027 parameter),\n since it is currently always set to skb-\u003elen. The code is updated with regard\n to this parameter change.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "a79ef76f4d8424324c2f108824a7398571193f43", "tree": "a80e84d19c867e1432fc05ae50199487ce2ea2b2", "parents": [ "841bac1d607d8bf2e068e4b24393fb77372814e3" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Tue Nov 28 19:51:42 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:31:01 2006 -0800" }, "message": "[DCCP] ccid3: Larger initial windows\n\nThis implements the larger-initial-windows feature for CCID 3, as described in\nsection 5 of RFC 4342. When the first feedback packet arrives, the sender can\nsend up to 2..4 packets per RTT, instead of just one.\n\nThe patch further\n * reduces the number of timestamping calls by passing the timestamp value\n (which is computed in one of the calling functions anyway) as argument\n\n * renames one constant with a very long name into one which is shorter and\n resembles the one in RFC 3448 (t_mbi)\n\n * simplifies some of the min_t/max_t cases where both `x\u0027, `y\u0027 have the same\n type\n\nCommiter note: renamed TFRC_t_mbi to TFRC_T_MBI, to follow Linux coding style.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "5aed324369c94a2c38469c8288e42eb1a9fac400", "tree": "76290075540058745b53d34db1672f5e292e6b1a", "parents": [ "78ad713da673a2977763521c347176137f3e493f" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Tue Nov 28 19:33:36 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:59 2006 -0800" }, "message": "[DCCP]: Tidy up unused structures\n\nThis removes and cleans up unused variables and structures which have become\nunnecessary following the introduction of the EWMA patch to automatically track\nthe CCID 3 receiver/sender packet sizes `s\u0027.\n\nIt deprecates the PACKET_SIZE socket option by returning an error code and\nprinting a deprecation warning if an application tries to read or write this\nsocket option.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "78ad713da673a2977763521c347176137f3e493f", "tree": "2ad817e9c90a46b8f0f2e41f60792fe5f00c865d", "parents": [ "2a1fda6f6c01d7ac195c040f14edcf9f64a5451e" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Tue Nov 28 19:22:33 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:58 2006 -0800" }, "message": "[DCCP] ccid3: Track RX/TX packet size `s\u0027 using moving-average\n\nProblem:\n" }, { "commit": "2a1fda6f6c01d7ac195c040f14edcf9f64a5451e", "tree": "81fb2bf85277d1f686bd484c89bf9fc4777a2396", "parents": [ "4384260443efe90a2ec0d907568dbc58ae792cd0" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Tue Nov 28 18:34:34 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:57 2006 -0800" }, "message": "[DCCP] ccid3: Set NoFeedback Timeout according to RFC 3448\n\nThis corrects the setting of the nofeedback timer with regard to RFC\n3448 - previously it was not set to max(4*R, 2*s/X) as specified. Using\nthe maximum of 1 second as upper bound (as it was done before) can have\ndetrimental effects, especially if R is small.\n\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "5d0dbc4a9b2d325458dcbf9a8329bd1d2cc7bd7e", "tree": "43aeec866470135e0b16420587ae04b7c41d43eb", "parents": [ "17893bc1a632e195574dc0dd9751243f0d5993d2" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 20:32:37 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:52 2006 -0800" }, "message": "[DCCP] ccid3: Consolidate handling of t_RTO\n\nThis patch\n * removes setting t_RTO in ccid3_hc_tx_init (per [RFC 3448, 4.2], t_RTO is\n undefined until feedback has been received);\n\n * makes some trivial changes (updates of comments);\n\n * performs a small optimisation by exploiting that the feedback timeout\n uses the value of t_ipi. The way it is done is safe, because the timeouts\n appear after the changes to t_ipi, ensuring that up-to-date values are used;\n\n * in ccid3_hc_tx_packet_recv, moves the t_rto statement closer to the calculation\n of the next_tmout. This makes the code clearer to read and is also safe, since\n t_rto is not updated until the next call of ccid3_hc_tx_packet_recv, and is not\n read by the functions called via ccid_wait_for_ccid();\n\n * removes a `max\u0027 statement in sk_reset_timer, this is not needed since the timeout\n value is always greater than 1E6 microseconds.\n\n * adds `XXX\u0027es to highlight that currently the nofeedback timer is set\n in a non-standard way\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "17893bc1a632e195574dc0dd9751243f0d5993d2", "tree": "7c036e23439d1cf5afbc63b147b7d1fbd8345eca", "parents": [ "48e03eee715b9e19df03153f2bcce6413632afcb" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 20:31:33 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:51 2006 -0800" }, "message": "[DCCP] ccid3: Consistently update t_nom, t_ipi, t_delta\n\nThis patch:\n\n * consolidates updating of parameters (t_nom, t_ipi, t_delta) which\n need to be updated at the same time, since they are inter-dependent\n\n * removes two inline functions which are no longer needed as a result of\n the above consolidation\n\n * resolves a FIXME regarding the re-calculation of t_ipi within the nofeedback\n timer, in the state where no feedback has previously been received\n\n * ties updating these parameters to updating the sending rate X, exploiting\n that all three parameters in turn depend on X; and using a small optimisation\n which can reduce the number of required instructions: only update the three\n parameters when X really changes\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "48e03eee715b9e19df03153f2bcce6413632afcb", "tree": "7a2995b3ac6bad9cac4309e97690dc0aa63a2868", "parents": [ "b798a9ede2b2434b9cbf0f3aee3e8598586fa89c" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 20:29:27 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:50 2006 -0800" }, "message": "[DCCP] ccid3: Consolidate timer resets\n\nThis patch concerns updating the value of the nofeedback timer when no feedback\nhas been received so far.\n\nSince in this case the value of R is still undefined according to [RFC 3448,\n4.2], we can not perform step (3) of [RFC 3448, 4.3]. A clarification is\nprovided in [RFC 4342, sec. 5], which states that in these cases the nofeedback\ntimer (still) expires \"after two seconds\".\n\nMany thanks to Ian McDonald for pointing this out and providing the\nclarification.\n\nThe patch\n * implements [RFC 4342, sec. 5] with regard to the above case\n * consolidates handling timer restart by\n\t- adding an appropriate jump label and\n\t- initialising the timeout value\n\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "5e19e3fcd7351de1ca87c4797cca27ba55c7e55e", "tree": "4d96f6c0c8fd8dc15a90c0172d433e4cb67ca3c3", "parents": [ "70dbd5b0ef3915f1e018e6437c8db9e999b0d701" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 12:28:48 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:41 2006 -0800" }, "message": "[DCCP] ccid3: Resolve small FIXME\n\nThis considers the case - ACK received while no packet has been sent\nso far. Resolved by printing a (rate-limited) warning message.\n\nFurther removes an unnecessary BUG_ON in ccid3_hc_tx_packet_recv,\nreceived feedback on a terminating connection is simply ignored.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "70dbd5b0ef3915f1e018e6437c8db9e999b0d701", "tree": "370f202cff866337ba3c2520bd33e9ef4ab17178", "parents": [ "da335baf9e788edfb00ee3b96f7b9526b6b2f8a9" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 12:27:55 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:40 2006 -0800" }, "message": "[DCCP] ccid3: Remove redundant statements in ccid3_hc_tx_packet_sent\n\nThis patch removes a switch statement which is redundant since,\n * nothing is done in states TFRC_SSTATE_NO_SENT/TFRC_SSTATE_NO_FBACK\n * it is impossible that the function is called in the state TFRC_SSTATE_TERM, since\n --the function is called, in dccp_write_xmit, after ccid3_hc_tx_send_packet\n --if ccid3_hc_tx_send_packet is called in state TFRC_SSTATE_TERM, it returns\n -EINVAL, which means that ccid3_hc_tx_packet_sent will not be called\n\t (compare dccp_write_xmit)\n --\u003e therefore, this case is logically impossible\n * the remaining state is TFRC_SSTATE_FBACK which conditionally updates t_ipi, t_nom,\n and t_delta. This is a no-op, since\n --t_ipi only changes when feedback is received\n --however, when feedback arrives via ccid3_hc_tx_packet_recv, there is an identical\n code block which performs the same set of operations\n --performing the same set of operations again in ccid3_hc_tx_packet_sent therefore\n does not change anything, since between the time of receiving the last feedback\n\t (and therefore update of t_ipi, t_nom, and t_delta), the value of t_ipi has not\n\t changed\n --since t_ipi has not changed, the values of t_delta and t_nom also do not change,\n they depend fully on t_ipi\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nAcked-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "da335baf9e788edfb00ee3b96f7b9526b6b2f8a9", "tree": "1eea49e2f94f359d9293025c64f381d43dd5922f", "parents": [ "7da7f456d7bc0e52009f882e8af0ac910293e157" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 12:26:57 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:39 2006 -0800" }, "message": "[DCCP] ccid3: Avoid congestion control on zero-sized data packets\n\nThis resolves an `XXX\u0027 in ccid3_hc_tx_send_packet().\n\nThe function is only called on Data and DataAck packets and returns a negative\nresult on zero-sized messages. This is a reasonable policy since CCID 3 is a\ncongestion-control module and congestion control on zero-sized Data(Ack)\npackets is in a way pathological.\n\nThe patch uses a more suitable error code for this case, it returns the Posix.1\ncode `EBADMSG\u0027 (\"Not a data message\") instead of `ENOTCONN\u0027.\n\nAs a result of ignoring zero-sized packets, a the condition for a warning\n\"First packet is data\" in ccid3_hc_tx_packet_sent is always satisfied; this\nmessage has been removed since it will always be printed.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "7da7f456d7bc0e52009f882e8af0ac910293e157", "tree": "c49cecb21030022b951fcdeed98728782f37d034", "parents": [ "91cf5a17257e1d2ef936fbf0223c3436ca583af9" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 12:26:03 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:38 2006 -0800" }, "message": "[DCCP] ccid3: Simplify control flow of ccid3_hc_tx_send_packet\n\nThis makes some logically equivalent simplifications, by replacing\nrc - values plus goto\u0027s with direct return statements.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "91cf5a17257e1d2ef936fbf0223c3436ca583af9", "tree": "eb7386ab5729d299bda6534c783378eb3fd73c4c", "parents": [ "f5c2d6367b04fd5ba98a5f9846b5fb870423968a" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 12:25:10 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:37 2006 -0800" }, "message": "[DCCP] ccid3: Fix calculation of t_ipi time of scheduled transmission\n\nProblem:\n" }, { "commit": "f5c2d6367b04fd5ba98a5f9846b5fb870423968a", "tree": "139687b3ec6b86b1938e7d55571880b9a88bf936", "parents": [ "90feeb951f61a80d3a8f8e5ced25b9ec78867eaf" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 12:22:48 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:36 2006 -0800" }, "message": "[DCCP] ccid3: Simplify control flow in the calculation of t_ipi\n\nThis patch performs a simplifying (performance) optimisation:\n\n In each call of the inline function ccid3_calc_new_t_ipi(), the state is\n tested against TFRC_SSTATE_NO_FBACK. This is expensive when the function\n is called very often. A simpler solution, implemented by this patch, is\n to adapt the control flow.\n\nBackground:\n" }, { "commit": "90feeb951f61a80d3a8f8e5ced25b9ec78867eaf", "tree": "da51cfafc387972099ab2540ae301f8ae3fd7e8c", "parents": [ "6472c051fcc5e571a9abee7f7a1ac58cc6e7bafa" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 27 12:13:38 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:35 2006 -0800" }, "message": "[DCCP] ccid3: Fix bug in calculation of first t_nom and first t_ipi\n\nProblem:\n" }, { "commit": "6472c051fcc5e571a9abee7f7a1ac58cc6e7bafa", "tree": "85060704f370434b82e29270ef17f67eb9ff22f7", "parents": [ "522f1d095bf76dbe2430fb9a9a257c0f27033f31" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Sun Nov 26 01:07:50 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:30:34 2006 -0800" }, "message": "[DCCP] ccid2: Allow window to grow larger\n\nNow that we can stuff bigger ack vectors into options.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "455431739ca2f4c7f02d0a5979559ac5a68a6f95", "tree": "b3ee30c13997b0411040f0641b1712f001a290b0", "parents": [ "ddfe10b82455a5a524055fc606d1f1562d388dc8" ], "author": { "name": "Ian McDonald", "email": "ian.mcdonald@jandi.co.nz", "time": "Mon Nov 20 18:44:03 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:24:41 2006 -0800" }, "message": "[DCCP] CCID3: Remove non-referenced variable\n\nThis removes a non-referenced variable.\n\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "23ea8945f6be2287fec67c85abcf24736c1ded80", "tree": "31f93549b53520710366561618dad7822f1c673f", "parents": [ "59348b19efebfd6a8d0791ff81d207b16594c94b" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 20 18:40:42 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:24:38 2006 -0800" }, "message": "[CCID 3]: Add annotations for socket structures\n\nThis adds documentation to the CCID 3 rx/tx socket fields, plus some\nminor re-formatting.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "59348b19efebfd6a8d0791ff81d207b16594c94b", "tree": "a9212a7bf62bd594cf02d23b9e33eb45a46d414d", "parents": [ "b1308dc015eb09cf094ca169296738a13ae049ad" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 20 18:39:23 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:24:38 2006 -0800" }, "message": "[DCCP]: Simplified conditions due to use of enum:8 states\n\nThis reaps the benefit of the earlier patch, which changed the type of\nCCID 3 states to use enums, in that many conditions are now simplified\nand the number of possible (unexpected) values is greatly reduced.\n\nIn a few instances, this also allowed to simplify pre-conditions; where\ncare has been taken to retain logical equivalence.\n\n[DCCP]: Introduce a consistent BUG/WARN message scheme\n\nThis refines the existing set of DCCP messages so that\n * BUG(), BUG_ON(), WARN_ON() have meaningful DCCP-specific counterparts\n * DCCP_CRIT (for severe warnings) is not rate-limited\n * DCCP_WARN() is introduced as rate-limited wrapper\n\nUsing these allows a faster and cleaner transition to their original\ncounterparts once the code has matured into a full DCCP implementation.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "56724aa434e9b4d73548021ede7a1474f533f3fe", "tree": "830f7c9883747a47cf63eddf05d8418e0f59a1cc", "parents": [ "84116716cc9404356f775443b460f76766f08f65" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 20 18:28:09 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:24:36 2006 -0800" }, "message": "[DCCP]: Add CCID3 debug support to Kconfig\n\nThis adds a CCID3 debug option to the configuration menu\nwhich is missing in Kconfig, but already used by the code.\n\nCCID 2 already provides such an entry.\n\nTo enable debugging, set CONFIG_IP_DCCP_CCID3_DEBUG\u003dy\n\nNOTE: The use of ccid3_{t,r}x_state_name is safe, since\n now only enum values can appear.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "84116716cc9404356f775443b460f76766f08f65", "tree": "627662e0ac97542de873ed2071561ba2c2f0ba8e", "parents": [ "f123bc5f800d528e59684c7faffda88b4dd2d38c" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Mon Nov 20 18:26:03 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:24:35 2006 -0800" }, "message": "[DCCP]: enable debug messages also for static builds\n\nThis patch\n * makes debugging (when configured) work both for static / module build\n * provides generic debugging macros for use in other DCCP / CCID modules\n * adds missing information about debug parameters to Kconfig\n * performs some code tidy-up\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "32aac18dfa0963fde40cc074ba97ebbae8b755f2", "tree": "c6851e77f1af7e7ef614e7e0a3bde630d57cf776", "parents": [ "58a5a7b9555ea231b557ebef5cabeaf8e951df0b" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Thu Nov 16 14:28:40 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:23:52 2006 -0800" }, "message": "[DCCP] CCID2: Code optimizations\n\nThese are code optimizations which are relevant when dealing with large\nwindows. They are not coded the way I would like to, but they do the job for\nthe short-term. This patch should be more neat.\n\nCommiter note: Changed the seqno comparisions to use {after,before}48 to handle\n wrapping.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "3c6952624a8f600f9a0fbc1f5db5560a7ef9b13e", "tree": "294924e950000cd786b0cdaf026ed58b2f81102d", "parents": [ "6bb100b9fc8f1ce330231b360028ab705a9f0378" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Wed Nov 15 21:27:47 2006 -0200" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Dec 02 21:23:49 2006 -0800" }, "message": "[DCCP]: Introduce DCCP_{BUG{_ON},CRIT} macros, use enum:8 for the ccid3 states\n\nThis patch tackles the following problem:\n * the ccid3_hc_{t,r}x_sock define ccid3hc{t,r}x_state as `u8\u0027, but\n in reality there can only be a few, pre-defined enum names\n * this necessitates addiditional checking for unexpected values\n which would otherwise be caught by the compiler\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "234af4840135342ab295b4e1219fd35c27fdd439", "tree": "cc19c28a68c181272148112444950d79efd95b81", "parents": [ "c8884edd078748905552d667857259e5358e1232" ], "author": { "name": "Randy Dunlap", "email": "randy.dunlap@oracle.com", "time": "Sun Oct 29 16:03:30 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Mon Oct 30 15:24:37 2006 -0800" }, "message": "[DCCP]: fix printk format warnings\n\nFix printk format warnings:\nbuild2.out:net/dccp/ccids/ccid2.c:355: warning: long long unsigned int format, u64 arg (arg 3)\nbuild2.out:net/dccp/ccids/ccid2.c:360: warning: long long unsigned int format, u64 arg (arg 3)\nbuild2.out:net/dccp/ccids/ccid2.c:482: warning: long long unsigned int format, u64 arg (arg 5)\nbuild2.out:net/dccp/ccids/ccid2.c:639: warning: long long unsigned int format, u64 arg (arg 3)\nbuild2.out:net/dccp/ccids/ccid2.c:639: warning: long long unsigned int format, u64 arg (arg 4)\nbuild2.out:net/dccp/ccids/ccid2.c:674: warning: long long unsigned int format, u64 arg (arg 3)\nbuild2.out:net/dccp/ccids/ccid2.c:720: warning: long long unsigned int format, u64 arg (arg 3)\n\nSigned-off-by: Randy Dunlap \u003crandy.dunlap@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0e64e94e477f8ed04e9295b11a5898d443c28a47", "tree": "0a02a3017d41a3a21038ff081b93b5b6359e1692", "parents": [ "977a415f2b70b5693aaa23b1a16ad57ea20a1dce" ], "author": { "name": "Gerrit Renker", "email": "gerrit@erg.abdn.ac.uk", "time": "Tue Oct 24 16:17:51 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue Oct 24 16:17:51 2006 -0700" }, "message": "[DCCP]: Update documentation references.\n\nUpdates the references to spec documents throughout the code, taking into\naccount that\n\n* the DCCP, CCID 2, and CCID 3 drafts all became RFCs in March this year\n\n* RFC 1063 was obsoleted by RFC 1191\n\n* draft-ietf-tcpimpl-pmtud-0x.txt was published as an Informational\n RFC, RFC 2923 on 2000-09-22.\n\nAll references verified.\n\nSigned-off-by: Gerrit Renker \u003cgerrit@erg.abdn.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "3dd9a7c3a155ee96160876cba92439fdc96d7e0b", "tree": "32c66cd4034d4c1f3ca7b7e7300a6f619a661c8a", "parents": [ "b83eff641ed39bd631535b9a8971e161b056f541" ], "author": { "name": "Ian McDonald", "email": "ian.mcdonald@jandi.co.nz", "time": "Fri Sep 22 14:26:44 2006 +1200" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Sep 24 18:03:41 2006 -0300" }, "message": "[DCCP]: Use constants for CCIDs\n\nWith constants for CCID numbers this now uses them in some places.\n\nSigned-off-by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "593f16aa627d61da447c76ee5a159450174627f6", "tree": "bad929ddf4eae76009c86e96e06cf31d1cd36a18", "parents": [ "374bcf32c86e1b56eab832bbb6b21e636707eab6" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Tue Sep 19 13:15:33 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:19:42 2006 -0700" }, "message": "[DCCP] CCID2: Add helper functions for changing important CCID2 state\n\nIntroduce methods which manipulate interesting congestion control\nstate such as pipe and rtt estimate. This is useful for people\nwishing to monitor the variables of CCID and instrument the code\n[perhaps using Kprobes]. Personally, I am a fan of\nencapsulation---that justifies this change \u003dD.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "374bcf32c86e1b56eab832bbb6b21e636707eab6", "tree": "c813985a0736740c683d841010b9317db0a2b1ee", "parents": [ "07978aabd52ce67f59971872c80f76d6e3ca18ae" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Tue Sep 19 13:14:43 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:19:41 2006 -0700" }, "message": "[DCCP] CCID2: Halve cwnd once upon multiple losses in a single RTT\n\nWhen multiple losses occur in one RTT, the window should be halved\nonly once [a single \"congestion event\"]. This is now implemented,\nalthough not perfectly. Slightly changed the interface for changing\nthe cwnd: pass hctx instead of dp. This is required in order to allow\nfor change_cwnd to be called from _init().\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "07978aabd52ce67f59971872c80f76d6e3ca18ae", "tree": "30b878674d4cf147cc6a0e91ff953298bc07b8e2", "parents": [ "8d424f6ca2d02026dadff409770639d720375afb" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Tue Sep 19 13:13:37 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:19:40 2006 -0700" }, "message": "[DCCP] CCID2: Allocate seq records on demand\n\nAllocate more sequence state on demand. Each time a packet is sent\nout by CCID2, a record of it needs to be kept. This list of records\ngrows proportionally to cwnd. Previously, the length of this list was\nhardcored and therefore the cwnd could only grow to this value (of\n128). Now, records are allocated on demand as necessary---cwnd may\ngrow as it wishes. The exceptional case of when memory is not\navailable is not handled gracefully. Perhaps, cwnd should be capped\nat that point.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8d424f6ca2d02026dadff409770639d720375afb", "tree": "5132c78f1e1aec21b9d264e416c490c4c7a119df", "parents": [ "446dec30c7f305ed1bb0092b0a8d9367d842a33f" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Tue Sep 19 13:12:44 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:19:39 2006 -0700" }, "message": "[DCCP] CCID2: Add Kconfig option for CCID2 debug\n\nAllow the user to choose whether or not to enable CCID2 debugging via\nKconfig.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "446dec30c7f305ed1bb0092b0a8d9367d842a33f", "tree": "91c7b8e94e973279974f55732b4b37426a3da2df", "parents": [ "69263bcfb5016bc3bdd099607a4232cba06f8491" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Tue Sep 19 13:10:11 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:19:39 2006 -0700" }, "message": "[DCCP] CCID2: Tell DCCP to quickly check whether cwnd is available\n\nIf not enough cwnd is available, tell the sender to check again as\nsoon as possible. This will increase CPU utilization (polling\nfrequently for cwnd) but will improve network performance. That is,\nthe sender will need to wait less before detecting the increase of\ncwnd. A better architecture would be for the CCID to call-back (or\ndequeue) from DCCP when it is able to transmit traffic -- not the\nother way around as it currently occurs.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "d458c25ce24ce00ea547e9976e293e7835416253", "tree": "b15e036537270aa0f70c390556aaa483702fd326", "parents": [ "29651cda97b0a9e4ac0fbeb5ea731a9909f0f128" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Tue Sep 19 13:07:20 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:19:11 2006 -0700" }, "message": "[DCCP] CCID2: Initialize ssthresh to infinity\n\nInitialize the slow-start threshold to infinity. This way, upon connection\ninitiation, slow-start will be exited only upon a packet loss. This patch will\nallow connections to quickly gain speed.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "29651cda97b0a9e4ac0fbeb5ea731a9909f0f128", "tree": "3b6002e4fa3bdf7d233b8d46654effe9f816d505", "parents": [ "4a0a50fb43912b4a593d2416c507a198fe607a6d" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Tue Sep 19 13:06:46 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:19:10 2006 -0700" }, "message": "[DCCP] CCID2: Fix jiffie wrap issues\n\nJiffies are now handled correctly (I hope) in CCID2. If they wrap, no\nproblem.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8e27e4650cb7e73aa4dd97d860539e7605ac7e39", "tree": "467ff3a8ecbb0fba014f917b42cc14f1f3665e32", "parents": [ "23d06e3b986677ec57007a24891fa9deb09ac973" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Tue Sep 19 13:05:35 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:19:08 2006 -0700" }, "message": "[DCCP] ackvec: Fix how DCCP_ACKVEC_STATE_NOT_RECEIVED is used\n\nFix the way state is masked out. DCCP_ACKVEC_STATE_NOT_RECEIVED is\ndefined as appears in the packet, therefore bit shifting is not\nrequired. This fix allows CCID2 to correctly detect losses.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "fc747e82b40ea50a62eb2aef55bedd4465607cb0", "tree": "31b98872a3dfe8f71740c7047ea5bd09c966e054", "parents": [ "8394e9b2faf539f82470b36c86f0485cab5278bd" ], "author": { "name": "Ian McDonald", "email": "ian.mcdonald@jandi.co.nz", "time": "Tue Aug 29 17:50:19 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Fri Sep 22 15:18:33 2006 -0700" }, "message": "[DCCP]: Tidyup CCID3 list handling\n\nAs Arnaldo Carvalho de Melo points out I should be using list_entry in case\nthe structure changes in future. Current code functions but is reliant\non position and requires type cast.\n\nNoticed when doing this that I have one more variable than I needed so\nremoving that also.\n\nSigned off by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "66a377c5041e1e399633153c8b500d457281e7c1", "tree": "9fa32d0504bf0a58181edb77940e0709f3f954a5", "parents": [ "3a13813e6effcfad5910d47b15b724621b50b878" ], "author": { "name": "Ian McDonald", "email": "ian.mcdonald@jandi.co.nz", "time": "Sat Aug 26 23:40:50 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Aug 26 23:40:50 2006 -0700" }, "message": "[DCCP]: Fix CCID3\n\nThis fixes CCID3 to give much closer performance to RFC4342.\n\nCCID3 is meant to alter sending rate based on RTT and loss.\n\nThe performance was verified against:\nhttp://wand.net.nz/~perry/max_download.php\n\nFor example I tested with netem and had the following parameters:\nDelayed Acks 1, MSS 256 bytes, RTT 105 ms, packet loss 5%.\n\nThis gives a theoretical speed of 71.9 Kbits/s. I measured across three\nruns with this patch set and got 70.1 Kbits/s. Without this patchset the\naverage was 232 Kbits/s which means Linux can\u0027t be used for CCID3 research\nproperly.\n\nI also tested with netem turned off so box just acting as router with 1.2\nmsec RTT. The performance with this is the same with or without the patch\nat around 30 Mbit/s.\n\nSigned off by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\n" }, { "commit": "80193aee18bc862e284ba18504f3a3e14706a997", "tree": "84b814bc136727976c9e9503093519a93b646529", "parents": [ "837d107cd101fbf734e9ea2bbb5c7336a329e432" ], "author": { "name": "Ian McDonald", "email": "ian.mcdonald@jandi.co.nz", "time": "Sat Aug 26 19:07:36 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Aug 26 19:07:36 2006 -0700" }, "message": "[DCCP]: Introduce dccp_rx_hist_find_entry\n\nThis adds a new function dccp_rx_hist_find_entry.\n\nSigned off by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "e6bccd357343e98db9e1fd0d487f4f924e1a7921", "tree": "21155efb0a977e9a408b0e6013b44bacff4b1881", "parents": [ "f3166c07175c1639687288006aeabed363a921f3" ], "author": { "name": "Ian McDonald", "email": "ian.mcdonald@jandi.co.nz", "time": "Sat Aug 26 19:01:30 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Aug 26 19:01:30 2006 -0700" }, "message": "[DCCP]: Update contact details and copyright\n\nJust updating copyright and contacts\n\nSigned off by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "f3166c07175c1639687288006aeabed363a921f3", "tree": "248bd80dbd38dea9a0eef2925588bb34ed4111f8", "parents": [ "59eed279c5daa88d95e429782ddb8ef87e52c44b" ], "author": { "name": "Ian McDonald", "email": "ian.mcdonald@jandi.co.nz", "time": "Sat Aug 26 19:01:03 2006 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sat Aug 26 19:01:03 2006 -0700" }, "message": "[DCCP]: Fix typo\n\nThis fixes a small typo in net/dccp/libs/packet_history.c\n\nSigned off by: Ian McDonald \u003cian.mcdonald@jandi.co.nz\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "6ab3d5624e172c553004ecc862bfeac16d9d68b7", "tree": "6d98881fe91fd9583c109208d5c27131b93fa248", "parents": [ "e02169b682bc448ccdc819dc8639ed34a23cedd8" ], "author": { "name": "Jörn Engel", "email": "joern@wohnheim.fh-wedel.de", "time": "Fri Jun 30 19:25:36 2006 +0200" }, "committer": { "name": "Adrian Bunk", "email": "bunk@stusta.de", "time": "Fri Jun 30 19:25:36 2006 +0200" }, "message": "Remove obsolete #include \u003clinux/config.h\u003e\n\nSigned-off-by: Jörn Engel \u003cjoern@wohnheim.fh-wedel.de\u003e\nSigned-off-by: Adrian Bunk \u003cbunk@stusta.de\u003e\n" }, { "commit": "2d0817d11eaec57435feb61493331a763f732a2b", "tree": "7c42229b1cc64e67efe8ed21b82cdc513e638f82", "parents": [ "110bae4efb5ed5565257a0fb9f6d26e6125a1c4b" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Mar 20 22:32:06 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 22:32:06 2006 -0800" }, "message": "[DCCP] options: Make dccp_insert_options \u0026 friends yell on error\n\nAnd not the silly LIMIT_NETDEBUG and silently return without inserting\nthe option requested.\n\nAlso drop some old debugging messages associated to option insertion.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c0c736db7ef4a7bdc1a28f3de751cc7e9f720313", "tree": "0935933828ed9021920cb5a922506911398c16ab", "parents": [ "45329e71eeb6cb483c6e7fe9e6f2849dd2932cb5" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Mar 20 22:05:37 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 22:05:37 2006 -0800" }, "message": "[DCCP] ccid2: coding style cleanups\n\nNo changes in the logic where made.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "72478873571d869906f7a250b09e12fa5b65e321", "tree": "562e6179922a5eef9c1f3eb7b1bf3813a6e53545", "parents": [ "c25a18ba347f091d1ce620ba33e6772b60a528e1" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Mar 20 22:00:37 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 22:00:37 2006 -0800" }, "message": "[DCCP] ipv6: Add missing ipv6 control socket\n\nI guess I forgot to add it, nah, now it just works:\n\n18:04:33.274066 IP6 ::1.1476 \u003e ::1.5001: request (service\u003d0)\n18:04:33.334482 IP6 ::1.5001 \u003e ::1.1476: reset (code\u003dbad_service_code)\n\nDitched IP_DCCP_UNLOAD_HACK, as now we would have to do it for both\nIPv6 and IPv4, so I\u0027ll come up with another way for freeing the\ncontrol sockets in upcoming changesets.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c25a18ba347f091d1ce620ba33e6772b60a528e1", "tree": "788c596356e150a26e0a67243ffc68213abdd493", "parents": [ "5e0817f84c33285c2ac7d3848e8896d025b5deff" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Mar 20 21:58:56 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 21:58:56 2006 -0800" }, "message": "[DCCP]: Uninline some functions\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "057fc6755a2381b4e81636659ef519e069f38623", "tree": "e15ad08e84e0892e1bb273067fd7e6f20a62fdad", "parents": [ "93ce20928f6e197707add8f670ae0cd029107e8f" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Mar 20 19:24:22 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 19:24:22 2006 -0800" }, "message": "[DCCP]: Kconfig tidy up\n\nMake CCID2 and CCID3 default to what was selected for DCCP and use the\nstandard short description for the CCIDs (TCP-Like \u0026 TCP-Friendly).\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "60fe62e789076ae7c13f7ffb35fec4b24802530d", "tree": "cc8d507a4276c3d5a2bcb72746660d485bc2e045", "parents": [ "a193a4abdd1f742a57f3f70b6a83c3e536876e97" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Mon Mar 20 19:23:32 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 19:23:32 2006 -0800" }, "message": "[DCCP]: sparse endianness annotations\n\nThis also fixes the layout of dccp_hdr short sequence numbers, problem\nwas not fatal now as we only support long (48 bits) sequence numbers.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "91f0ebf7b6d5cb2b6e818d48587566144821babe", "tree": "505c66f36bd72014d7eacb7a04ea011bae2e9a3a", "parents": [ "f38c39d6ce8226519455a6dfe91c2ad84f363f6f" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Mar 20 19:21:44 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 19:21:44 2006 -0800" }, "message": "[DCCP] CCID: Improve CCID infrastructure\n\n1. No need for -\u003eccid_init nor -\u003eccid_exit, this is what module_{init,exit}\n does and anynways neither ccid2 nor ccid3 were using it.\n\n2. Rename struct ccid to struct ccid_operations and introduce struct ccid\n with a pointer to ccid_operations and rigth after it the rx or tx\n private state.\n\n3. Remove the pointer to the state of the half connections from struct\n dccp_sock, now its derived thru ccid_priv() from the ccid pointer.\n\nNow we also can implement the setsockopt for changing the CCID easily as\nno ccid init routines can affect struct dccp_sock in any way that prevents\nother CCIDs from working if a CCID switch operation is asked by apps.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "77ff72d528d5b9d30a47f42f364ba34d931f9da3", "tree": "a222cdf545a9156da307d075f4645a235c030596", "parents": [ "ba66c6e8b292997467128506f39fa6607e959050" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Mon Mar 20 17:57:52 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 17:57:52 2006 -0800" }, "message": "[DCCP] CCID2: Drop sock reference count on timer expiration and reset.\n\nThere was a hybrid use of standard timers and sk_timers. This caused\nthe reference count of the sock to be incorrect when resetting the RTO\ntimer. The sock reference count should now be correct, enabling its\ndestruction, and allowing the DCCP module to be unloaded.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "2a91aa3967398fb94eccc8da67c82bce9f67afdf", "tree": "62bf003487121bc629919c85810df11e52016b8f", "parents": [ "aa5d7df3b20e0e493e90e1151510ab3ae8366bb5" ], "author": { "name": "Andrea Bittau", "email": "a.bittau@cs.ucl.ac.uk", "time": "Mon Mar 20 17:41:47 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 17:41:47 2006 -0800" }, "message": "[DCCP] CCID2: Initial CCID2 (TCP-Like) implementation\n\nOriginal work by Andrea Bittau, Arnaldo Melo cleaned up and fixed several\nissues on the merge process.\n\nFor now CCID2 was turned the default for all SOCK_DCCP connections, but this\nwill be remedied soon with the merge of the feature negotiation code.\n\nSigned-off-by: Andrea Bittau \u003ca.bittau@cs.ucl.ac.uk\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "aa5d7df3b20e0e493e90e1151510ab3ae8366bb5", "tree": "1decadd46dd3d3dd88c9e812c46293398a79d1ab", "parents": [ "9833d6da00c95e8a471411fb079da6b25787b05e" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Mar 20 17:35:13 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 17:35:13 2006 -0800" }, "message": "[DCCP] CCID3: Set the no_feedback_timer fields near init_timer\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "411447019ab583c659600b0519db5658a2444f11", "tree": "3bf3e7c655c91097cead92c65334a4ebbb51906e", "parents": [ "02bcf28c82c8e4b72c4b89bddbbb6fea1a646d07" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Mar 20 17:20:23 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 20 17:20:23 2006 -0800" }, "message": "[DCCP] CCID: Allow ccid_{init,exit} to be NULL\n\nTesting if the ccid being instantiated has these methods in\nccid_init().\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "c09966608da7e8cad3468e925ac9062e44fee831", "tree": "4e3b04e3aabe5bf5106fe04913ad7b75525d7de4", "parents": [ "c027f5f995d8b6efc934be384085e3e8425638e4" ], "author": { "name": "Ian McDonald", "email": "imcdnzl@gmail.com", "time": "Fri Mar 03 17:54:46 2006 -0800" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Sat Mar 04 21:06:29 2006 -0800" }, "message": "[DCCP] ccid3: Divide by zero fix\n\nIn rare circumstances 0 is returned by dccp_li_hist_calc_i_mean which\nleads to a divide by zero in ccid3_hc_rx_packet_recv. Explicitly check\nfor zero return now. Update copyright notice at same time.\n\nFound by Arnaldo.\n\nSigned-off-by: Ian McDonald \u003cimcdnzl@gmail.com\u003e\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "1b8623545b42c03eb92e51b28c84acf4b8ba00a3", "tree": "071045ad9c60d2697292c523c77322a70a248fb9", "parents": [ "92118c739df879497b8cc5a2eb3a9dc255f01b20" ], "author": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Thu Dec 15 01:07:03 2005 -0500" }, "committer": { "name": "Al Viro", "email": "viro@zeniv.linux.org.uk", "time": "Tue Feb 07 20:56:35 2006 -0500" }, "message": "[PATCH] remove bogus asm/bug.h includes.\n\nA bunch of asm/bug.h includes are both not needed (since it will get\npulled anyway) and bogus (since they are done too early). Removed.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "dd0fc66fb33cd610bc1a5db8a5e232d34879b4d7", "tree": "51f96a9db96293b352e358f66032e1f4ff79fafb", "parents": [ "3b0e77bd144203a507eb191f7117d2c5004ea1de" ], "author": { "name": "Al Viro", "email": "viro@ftp.linux.org.uk", "time": "Fri Oct 07 07:46:04 2005 +0100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Oct 08 15:00:57 2005 -0700" }, "message": "[PATCH] gfp flags annotations - part 1\n\n - added typedef unsigned int __nocast gfp_t;\n\n - replaced __nocast uses for gfp flags with gfp_t - it gives exactly\n the same warnings as far as sparse is concerned, doesn\u0027t change\n generated code (from gcc point of view we replaced unsigned int with\n typedef) and documents what\u0027s going on far better.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "88f964db6ef728982734356bf4c406270ea29c1d", "tree": "7fb9ba2fb646f4917911fed4a0a37cd52a12eae4", "parents": [ "561713cf475de1f671cc89c437927ec008a20209" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Sep 18 00:19:32 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Sep 18 00:19:32 2005 -0700" }, "message": "[DCCP]: Introduce CCID getsockopt for the CCIDs\n\nAllocation for the optnames is similar to the DCCP options, with a\nrange for rx and tx half connection CCIDs.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "65299d6c3cfb49cc3eee4fc483e7edd23ea7b2ed", "tree": "111ec511694c75d2bc0f7404f6ea02b6230cae7a", "parents": [ "ae31c3399d17b1f7bc1742724f70476b5417744f" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Sun Sep 18 00:18:32 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Sep 18 00:18:32 2005 -0700" }, "message": "[CCID3]: Introduce include/linux/tfrc.h\n\nMoving the TFRC sender and receiver variables to separate structs, so\nthat we can copy these structs to userspace thru getsockopt,\ndccp_diag, etc.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "59c2353dd045509626d7cc1953cb2d6b7358ab7a", "tree": "d5afb3cc1e015cea8f2bc7506cadd2d7457f8e36", "parents": [ "121caf577dc61588c1128821873a08cec48baf58" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Mon Sep 12 14:16:58 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Sep 12 14:16:58 2005 -0700" }, "message": "[CCID3]: Listen socks doesn\u0027t have a private CCID block\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "59d203f9e989c81defc2cb1c559d678e16dbe684", "tree": "bd8456b28612818674b5031b8c0dcaa752b324bd", "parents": [ "dc19336c76d213fd9ae7b6a352e837f509418012" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 20:01:25 2005 -0300" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 20:01:25 2005 -0300" }, "message": "[CCID3] Cleanup ccid3 debug calls\n\nAlso use some BUG_ON where appropriate and use LIMIT_NETDEBUG for the unlikely\ncases where we, at this stage, want to know about, that in my tests hasn\u0027t\nappeared in the radar.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "d7e0fb985cb033e605c3167ff183b5e98b7f7644", "tree": "42e966c189f4ffef2321c3ae7483b862e4c5102e", "parents": [ "cb7b593c2c808b32a1ea188599713c434b95f849" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 19:58:18 2005 -0300" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 19:58:18 2005 -0300" }, "message": "[CCID3] Initialize ccid3hctx_t_ipi to 250ms\n\nTo match more closely what is described in RFC 3448.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\nSigned-off-by: Ian McDonald \u003ciam4@cs.waikato.ac.nz\u003e\n" }, { "commit": "59725dc2a2e86a03bbf97976ede3bdc6f95bba92", "tree": "2c529321fc918a7d8c7affddd35227a497e082f0", "parents": [ "b0e567806d16586629468c824dfb2e71155df7da" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:40:58 2005 -0300" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:40:58 2005 -0300" }, "message": "[CCID3] Introduce ccid3_hc_[rt]x_sk() for overal consistency\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "b0e567806d16586629468c824dfb2e71155df7da", "tree": "70f5905fae3270985fed1050335c56598fd32646", "parents": [ "954ee31f366fabe53fb450482789258fe552f40b" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:38:35 2005 -0300" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:38:35 2005 -0300" }, "message": "[DCCP] Introduce dccp_timestamp\n\nTo start the timestamps with 0.0ms, easing the integer maths in the CCIDs, this\nprobably will be reworked to use the to be introduced struct timeval_offset\ninfrastructure out of skb_get_timestamp, etc.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "954ee31f366fabe53fb450482789258fe552f40b", "tree": "9118ac57a87aefdbb20a02366cc96a70c9ca2b10", "parents": [ "b3a3077d963fc54a25be26e2e84fe9f4327c1e12" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:37:05 2005 -0300" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:37:05 2005 -0300" }, "message": "[CCID3] Initialize more fields in ccid3_hc_rx_init\n\nThe initialization of ccid3hcrx_rtt to 5ms is just a bandaid, I\u0027ll continue\nauditing the CCID3 HC rx codebase to fix this properly, probably I\u0027ll add a\nfeedback timer as suggested in the CCID3 draft.\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "b3a3077d963fc54a25be26e2e84fe9f4327c1e12", "tree": "5f7c940c3f0758a130d3a6a3492a7a69f8758e53", "parents": [ "1a28599a2c2781dd6af72f4f84175e2db74d3bb1" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:34:10 2005 -0300" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:34:10 2005 -0300" }, "message": "[CCID3] Make the ccid3hcrx_rtt calc look more like the ccid3hctx_rtt one\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" }, { "commit": "1a28599a2c2781dd6af72f4f84175e2db74d3bb1", "tree": "1cbb3911a234da8722bbd87445ca5b623d9328c8", "parents": [ "1c14ac0ae8eb62cbb40af1e31b156994c7d7d3d5" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:32:56 2005 -0300" }, "committer": { "name": "Arnaldo Carvalho de Melo", "email": "acme@mandriva.com", "time": "Fri Sep 09 02:32:56 2005 -0300" }, "message": "[CCID3] Use ELAPSED_TIME in the HC TX RTT estimation\n\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@mandriva.com\u003e\n" } ], "next": "27ae543e6f116df66e2b19ff0a3aa1053e4784d8" }