aboutsummaryrefslogtreecommitdiffstats
BranchCommit messageAuthorAge
archive/unc-master-3.0P-FP: fix BUG_ON releated to priority inheritanceBjoern Brandenburg13 years
archived-2013.1uncachedev: mmap memory that is not cached by CPUsGlenn Elliott12 years
archived-private-masterMerge branch 'wip-2.6.34' into old-private-masterAndrea Bastoni15 years
archived-semi-partMerge branch 'wip-semi-part' of ssh://cvs/cvs/proj/litmus/repo/litmus2010 int...Andrea Bastoni15 years
demoFurther refinementsJonathan Herman14 years
ecrts-pgm-finalMerge branch 'wip-ecrts14-pgm' of ssh://rtsrv.cs.unc.edu/home/litmus/litmus-r...Glenn Elliott12 years
ecrts14-pgm-finalMerge branch 'wip-ecrts14-pgm' of ssh://rtsrv.cs.unc.edu/home/litmus/litmus-r...Glenn Elliott12 years
gpusync-rtss12Final GPUSync implementation.Glenn Elliott12 years
gpusync/stagingRename IKGLP R2DGLP.Glenn Elliott12 years
linux-tipMerge branch 'slab/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/p...Linus Torvalds15 years
litmus2008-patch-seriesadd i386 feather-trace implementationBjoern B. Brandenburg16 years
masterPSN-EDF: use inferred_sporadic_job_release_atBjoern Brandenburg9 years
pgmmake it compileGlenn Elliott12 years
prop/litmus-signalsInfrastructure for Litmus signals.Glenn Elliott13 years
prop/robust-tie-breakFixed bug in edf_higher_prio().Glenn Elliott13 years
stagingFix tracepoint compilation errorFelipe Cerqueira13 years
test9/23/2016Namhoon Kim9 years
tracing-develTest kernel tracing events capabilitiesAndrea Bastoni16 years
v2.6.34-with-arm-patchessmsc911x: Add spinlocks around registers accessCatalin Marinas15 years
v2015.1Add ARM syscall def for get_current_budgetBjoern Brandenburg10 years
wip-2011.2-bbbLitmus core: simplify np-section protocolBjoern B. Brandenburg14 years
wip-2011.2-bbb-traceRefactor sched_trace_log_message() -> debug_trace_log_message()Andrea Bastoni14 years
wip-2012.3-gpuSOBLIV draining support for C-EDF.Glenn Elliott12 years
wip-2012.3-gpu-preportpick up last C-RM fileGlenn Elliott12 years
wip-2012.3-gpu-rtss13Fix critical bug in GPU tracker.Glenn Elliott12 years
wip-2012.3-gpu-sobliv-budget-w-ksharkProper sobliv draining and many bug fixes.Glenn Elliott12 years
wip-aedzl-finalMake it easier to compile AEDZL interfaces in liblitmus.Glenn Elliott15 years
wip-aedzl-revisedAdd sched_trace data for Apative EDZLGlenn Elliott15 years
wip-arbit-deadlineFix compilation bug.Glenn Elliott13 years
wip-aux-tasksDescription of refined aux task inheritance.Glenn Elliott13 years
wip-bbbGSN-EDF & Core: improve debug TRACE'ing for NP sectionsBjoern B. Brandenburg14 years
wip-bbb-prio-donuse correct timestampBjoern B. Brandenburg14 years
wip-better-breakImplement hash-based EDF tie-breaking.Glenn Elliott13 years
wip-binary-heapMake C-EDF work with simplified binheap_deleteGlenn Elliott13 years
wip-budgetAdded support for choices in budget policy enforcement.Glenn Elliott15 years
wip-colorSummarize schedulability with final recordJonathan Herman13 years
wip-color-jlhsched_color: Fixed two bugs causing crashing on experiment restart and a rare...Jonathan Herman13 years
wip-d10-hz1000Enable HZ=1000 on District 10Bjoern B. Brandenburg15 years
wip-default-clusteringFeature: Make default C-EDF clustering compile-time configurable.Glenn Elliott15 years
wip-dissipation-jericksoUpdate from 2.6.36 to 2.6.36.4Jeremy Erickson11 years
wip-dissipation2-jericksoUpdate 2.6.36 to 2.6.36.4Jeremy Erickson11 years
wip-ecrts14-pgmMerge branch 'wip-ecrts14-pgm' of ssh://rtsrv.cs.unc.edu/home/litmus/litmus-r...Glenn Elliott12 years
wip-edf-hsblast tested versionJonathan Herman14 years
wip-edf-osLookup table EDF-osJeremy Erickson12 years
wip-edf-tie-breakMerge branch 'wip-edf-tie-break' of ssh://rtsrv.cs.unc.edu/home/litmus/litmus...Glenn Elliott13 years
wip-edzl-critiqueUse hr_timer's active checks instead of having own flag.Glenn Elliott15 years
wip-edzl-finalImplementation of the EDZL scheduler.Glenn Elliott15 years
wip-edzl-revisedClean up comments.Glenn Elliott15 years
wip-eventsAdded support for tracing arbitrary actions.Jonathan Herman15 years
wip-extra-debugDBG: add additional tracingBjoern B. Brandenburg15 years
wip-fix-switch-jericksoAttempt to fix race condition with plugin switchingJeremy Erickson15 years
wip-fix3sched: show length of runqueue clock deactivation in /proc/sched_debugBjoern B. Brandenburg15 years
wip-fmlp-dequeueImprove FMLP queue management.Glenn Elliott14 years
wip-ft-irq-flagFeather-Trace: keep track of interrupt-related interference.Bjoern B. Brandenburg14 years
wip-gpu-cleanupEnable sched_trace log injection from userspaceGlenn Elliott13 years
wip-gpu-interruptsRemove option for threading of all softirqs.Glenn Elliott14 years
wip-gpu-rtas12Generalized GPU cost predictors + EWMA. (untested)Glenn Elliott13 years
wip-gpu-rtss12Final GPUSync implementation.Glenn Elliott13 years
wip-gpu-rtss12-srpexperimental changes to support GPUs under SRPGlenn Elliott13 years
wip-gpusync-mergeCleanup priority tracking for budget enforcement.Glenn Elliott11 years
wip-ikglpMove RSM and IKGLP imp. to own .c filesGlenn Elliott13 years
wip-k-fmlpMerge branch 'mpi-master' into wip-k-fmlpGlenn Elliott14 years
wip-kernel-coloringAdded recolor syscallNamhoon Kim7 years
wip-kernthreadsKludge work-queue processing into klitirqd.Glenn Elliott15 years
wip-klmirqd-to-auxAllow klmirqd threads to be given names.Glenn Elliott13 years
wip-ksharkMerge branch 'mpi-staging' into wip-ksharkJonathan Herman13 years
wip-litmus-3.2Merge commit 'v3.2' into litmus-stagingAndrea Bastoni13 years
wip-litmus2011.2Cleanup: Coding conformance for affinity stuff.Glenn Elliott14 years
wip-litmus3.0-2011.2Feather-Trace: keep track of interrupt-related interference.Bjoern B. Brandenburg14 years
wip-master-2.6.33-rtAvoid deadlock when switching task policy to BACKGROUND (ugly)Andrea Bastoni15 years
wip-mcRemoved ARM-specific hacks which disabled less common mixed-criticality featu...Jonathan Herman12 years
wip-mc-bipasaMC-EDF addedbipasa chattopadhyay13 years
wip-mc-jericksoSplit C/D queuesJeremy Erickson15 years
wip-mc2-cache-slackManually patched mc^2 related codeMing Yang10 years
wip-mcrit-maccosmeticMac Mollison15 years
wip-merge-3.0Prevent Linux to send IPI and queue tasks on remote CPUs.Andrea Bastoni14 years
wip-merge-v3.0Prevent Linux to send IPI and queue tasks on remote CPUs.Andrea Bastoni14 years
wip-migration-affinityNULL affinity dereference in C-EDF.Glenn Elliott14 years
wip-mmap-uncacheshare branch with othersGlenn Elliott13 years
wip-modechangeRTSS 2017 submissionNamhoon Kim8 years
wip-nested-lockingAppears to be working.Bryan Ward12 years
wip-omlp-gedfFirst implementation of G-OMLP.Glenn Elliott15 years
wip-paiSome cleanup of PAIGlenn Elliott14 years
wip-percore-lib9/21/2016Namhoon Kim9 years
wip-performanceCONFIG_DONT_PREEMPT_ON_TIE: Don't preeempt a scheduled task on priority tie.Glenn Elliott14 years
wip-pgmAdd PGM support to C-FLGlenn Elliott12 years
wip-pgm-splitFirst draft of C-FL-splitNamhoon Kim12 years
wip-pm-ovdAdd preemption-and-migration overhead tracing supportAndrea Bastoni15 years
wip-prio-inhP-EDF updated to use the generic pi framework.Glenn Elliott15 years
wip-prioq-dglBUG FIX: Support DGLs with PRIOQ_MUTEXGlenn Elliott13 years
wip-refactored-gedfGeneralizd architecture for GEDF-style scheduelrs to reduce code redundancy.Glenn Elliott15 years
wip-release-master-fixbugfix: release master CPU must signal task was pickedBjoern B. Brandenburg14 years
wip-robust-tie-breakEDF priority tie-breaks.Glenn Elliott13 years
wip-rt-ksharkMove task time accounting into the complete_job method.Jonathan Herman13 years
wip-rtas12-pgmScheduling of PGM jobs.Glenn Elliott13 years
wip-semi-partFix compile error with newer GCCJeremy Erickson12 years
wip-semi-part-edfos-jericksoUse initial CPU set by clientJeremy Erickson12 years
wip-shared-libTODO: Fix condition checks in replicate_page_move_mapping()Namhoon Kim9 years
wip-shared-lib2RTAS 2017 Submission ver.Namhoon Kim9 years
wip-shared-memInitial commit for shared libraryNamhoon Kim9 years
wip-splitting-jericksoFix release behaviorJeremy Erickson13 years
wip-splitting-omlp-jericksoBjoern's Dissertation Code with Priority DonationJeremy Erickson13 years
wip-stage-binheapAn efficient binary heap implementation.Glenn Elliott13 years
wip-sun-portDynamic memory allocation and clean exit for FeatherTraceChristopher Kenna15 years
wip-timer-tracebugfix: C-EDF, clear scheduled field of the correct CPU upon task_exitAndrea Bastoni15 years
wip-tracepointsAdd kernel-style events for sched_trace_XXX() functionsAndrea Bastoni14 years
 
TagDownloadAuthorAge
2015.1commit 8e51b37822...Bjoern Brandenburg10 years
2013.1commit bcaacec1ca...Glenn Elliott12 years
2012.3commit c158b5fbe4...Jonathan Herman13 years
2012.2commit b53c479a0f...Glenn Elliott13 years
2012.1commit 83b11ea1c6...Bjoern B. Brandenburg14 years
rtas12-mc-beta-expcommit 8e236ee20f...Christopher Kenna14 years
2011.1commit d11808b5c6...Christopher Kenna15 years
v2.6.37-rc4commit e8a7e48bb2...Linus Torvalds15 years
v2.6.37-rc3commit 3561d43fd2...Linus Torvalds15 years
v2.6.37-rc2commit e53beacd23...Linus Torvalds15 years
v2.6.37-rc1commit c8ddb2713c...Linus Torvalds15 years
v2.6.36commit f6f94e2ab1...Linus Torvalds15 years
2010.2commit 5c5456402d...Bjoern B. Brandenburg15 years
v2.6.36-rc8commit cd07202cc8...Linus Torvalds15 years
v2.6.36-rc7commit cb655d0f3d...Linus Torvalds15 years
v2.6.36-rc6commit 899611ee7d...Linus Torvalds15 years
v2.6.36-rc5commit b30a3f6257...Linus Torvalds15 years
v2.6.36-rc4commit 49553c2ef8...Linus Torvalds15 years
v2.6.36-rc3commit 2bfc96a127...Linus Torvalds15 years
v2.6.36-rc2commit 76be97c1fc...Linus Torvalds15 years
v2.6.36-rc1commit da5cabf80e...Linus Torvalds15 years
v2.6.35commit 9fe6206f40...Linus Torvalds15 years
v2.6.35-rc6commit b37fa16e78...Linus Torvalds15 years
v2.6.35-rc5commit 1c5474a65b...Linus Torvalds15 years
v2.6.35-rc4commit 815c4163b6...Linus Torvalds15 years
v2.6.35-rc3commit 7e27d6e778...Linus Torvalds15 years
v2.6.35-rc2commit e44a21b726...Linus Torvalds15 years
v2.6.35-rc1commit 67a3e12b05...Linus Torvalds15 years
2010.1commit 7c1ff4c544...Andrea Bastoni15 years
v2.6.34commit e40152ee1e...Linus Torvalds15 years
v2.6.33.4commit 4640b4e7d9...Greg Kroah-Hartman15 years
v2.6.34-rc7commit b57f95a382...Linus Torvalds15 years
v2.6.34-rc6commit 66f41d4c5c...Linus Torvalds15 years
v2.6.33.3commit 3e7ad8ed97...Greg Kroah-Hartman15 years
v2.6.34-rc5commit 01bf0b6457...Linus Torvalds15 years
v2.6.34-rc4commit 0d0fb0f9c5...Linus Torvalds15 years
v2.6.33.2commit 19f00f070c...Greg Kroah-Hartman15 years
v2.6.34-rc3commit 2eaa9cfdf3...Linus Torvalds15 years
v2.6.34-rc2commit 220bf991b0...Linus Torvalds15 years
v2.6.33.1commit dbdafe5ccf...Greg Kroah-Hartman15 years
v2.6.34-rc1commit 57d54889cd...Linus Torvalds16 years
v2.6.33commit 60b341b778...Linus Torvalds16 years
v2.6.33-rc8commit 724e6d3fe8...Linus Torvalds16 years
v2.6.33-rc7commit 29275254ca...Linus Torvalds16 years
v2.6.33-rc6commit abe94c756c...Linus Torvalds16 years
v2.6.33-rc5commit 92dcffb916...Linus Torvalds16 years
v2.6.33-rc4commit 7284ce6c9f...Linus Torvalds16 years
v2.6.33-rc3commit 74d2e4f8d7...Linus Torvalds16 years
v2.6.33-rc2commit 6b7b284958...Linus Torvalds16 years
v2.6.33-rc1commit 55639353a0...Linus Torvalds16 years
v2.6.32commit 22763c5cf3...Linus Torvalds16 years
v2.6.32-rc8commit 648f4e3e50...Linus Torvalds16 years
v2.6.32-rc7commit 156171c71a...Linus Torvalds16 years
v2.6.32-rc6commit b419148e56...Linus Torvalds16 years
v2.6.32-rc5commit 012abeea66...Linus Torvalds16 years
v2.6.32-rc4commit 161291396e...Linus Torvalds16 years
v2.6.32-rc3commit 374576a8b6...Linus Torvalds16 years
v2.6.32-rc1commit 17d857be64...Linus Torvalds16 years
v2.6.32-rc2commit 17d857be64...Linus Torvalds16 years
v2.6.31commit 74fca6a428...Linus Torvalds16 years
v2.6.31-rc9commit e07cccf404...Linus Torvalds16 years
v2.6.31-rc8commit 326ba5010a...Linus Torvalds16 years
v2.6.31-rc7commit 422bef879e...Linus Torvalds16 years
v2.6.31-rc6commit 64f1607ffb...Linus Torvalds16 years
v2.6.31-rc5commit ed680c4ad4...Linus Torvalds16 years
v2.6.31-rc4commit 4be3bd7849...Linus Torvalds16 years
v2.6.31-rc3commit 6847e154e3...Linus Torvalds16 years
v2.6.31-rc2commit 8e4a718ff3...Linus Torvalds16 years
v2.6.31-rc1commit 28d0325ce6...Linus Torvalds16 years
v2.6.30commit 07a2039b8e...Linus Torvalds16 years
v2.6.30-rc8commit 9fa7eb283c...Linus Torvalds16 years
v2.6.30-rc7commit 59a3759d0f...Linus Torvalds16 years
v2.6.30-rc6commit 1406de8e11...Linus Torvalds16 years
v2.6.30-rc5commit 091bf7624d...Linus Torvalds16 years
v2.6.30-rc4commit 091438dd56...Linus Torvalds16 years
v2.6.30-rc3commit 0910697403...Linus Torvalds16 years
v2.6.30-rc2commit 0882e8dd3a...Linus Torvalds16 years
v2.6.30-rc1commit 577c9c456f...Linus Torvalds16 years
v2.6.29commit 8e0ee43bc2...Linus Torvalds16 years
v2.6.29-rc8commit 041b62374c...Linus Torvalds16 years
v2.6.29-rc7commit fec6c6fec3...Linus Torvalds17 years
v2.6.29-rc6commit 20f4d6c3a2...Linus Torvalds17 years
v2.6.29-rc5commit d2f8d7ee1a...Linus Torvalds17 years
v2.6.29-rc4commit 8e4921515c...Linus Torvalds17 years
v2.6.29-rc3commit 18e352e4a7...Linus Torvalds17 years
v2.6.29-rc2commit 1de9e8e70f...Linus Torvalds17 years
v2.6.29-rc1commit c59765042f...Linus Torvalds17 years
v2.6.28commit 4a6908a3a0...Linus Torvalds17 years
v2.6.28-rc9commit 929096fe9f...Linus Torvalds17 years
v2.6.28-rc8commit 8b1fae4e42...Linus Torvalds17 years
v2.6.28-rc7commit 061e41fdb5...Linus Torvalds17 years
v2.6.28-rc6commit 13d428afc0...Linus Torvalds17 years
v2.6.28-rc5commit 9bf1a2445f...Linus Torvalds17 years
v2.6.28-rc4commit f7160c7573...Linus Torvalds17 years
v2.6.28-rc3commit 45beca08dd...Linus Torvalds17 years
v2.6.28-rc2commit 0173a3265b...Linus Torvalds17 years
v2.6.28-rc1commit 57f8f7b60d...Linus Torvalds17 years
v2.6.27commit 3fa8749e58...Linus Torvalds17 years
v2.6.27-rc9commit 4330ed8ed4...Linus Torvalds17 years
v2.6.27-rc8commit 94aca1dac6...Linus Torvalds17 years
v2.6.27-rc7commit 72d31053f6...Linus Torvalds17 years
v2.6.27-rc6commit adee14b2e1...Linus Torvalds17 years
v2.6.27-rc5commit 24342c34a0...Linus Torvalds17 years
v2.6.27-rc4commit 6a55617ed5...Linus Torvalds17 years
v2.6.27-rc3commit 30a2f3c60a...Linus Torvalds17 years
v2.6.27-rc2commit 0967d61ea0...Linus Torvalds17 years
v2.6.27-rc1commit 6e86841d05...Linus Torvalds17 years
v2.6.26commit bce7f793da...Linus Torvalds17 years
v2.6.26-rc9commit b7279469d6...Linus Torvalds17 years
v2.6.26-rc8commit 543cf4cb3f...Linus Torvalds17 years
v2.6.26-rc7commit d70ac829b7...Linus Torvalds17 years
v2.6.26-rc6commit 5dd34572ad...Linus Torvalds17 years
v2.6.26-rc5commit 53c8ba9540...Linus Torvalds17 years
v2.6.26-rc4commit e490517a03...Linus Torvalds17 years
v2.6.26-rc3commit b8291ad07a...Linus Torvalds17 years
v2.6.26-rc2commit 492c2e476e...Linus Torvalds17 years
v2.6.26-rc1commit 2ddcca36c8...Linus Torvalds17 years
v2.6.25commit 4b119e21d0...Linus Torvalds17 years
v2.6.25-rc9commit 120dd64cac...Linus Torvalds17 years
v2.6.25-rc8commit 0e81a8ae37...Linus Torvalds17 years
v2.6.25-rc7commit 05dda977f2...Linus Torvalds17 years
v2.6.25-rc6commit a978b30af3...Linus Torvalds17 years
v2.6.25-rc5commit cdeeeae056...Linus Torvalds18 years
v2.6.25-rc4commit 29e8c3c304...Linus Torvalds18 years
v2.6.25-rc3commit bfa274e243...Linus Torvalds18 years
v2.6.25-rc2commit 101142c37b...Linus Torvalds18 years
v2.6.25-rc1commit 19af35546d...Linus Torvalds18 years
v2.6.24commit 49914084e7...Linus Torvalds18 years
v2.6.24-rc8commit cbd9c88369...Linus Torvalds18 years
v2.6.24-rc7commit 3ce5445046...Linus Torvalds18 years
v2.6.24-rc6commit ea67db4cdb...Linus Torvalds18 years
v2.6.24-rc5commit 82d29bf6dc...Linus Torvalds18 years
v2.6.24-rc4commit 09b56adc98...Linus Torvalds18 years
v2.6.24-rc3commit d9f8bcbf67...Linus Torvalds18 years
v2.6.24-rc2commit dbeeb816e8...Linus Torvalds18 years
v2.6.24-rc1commit c9927c2bf4...Linus Torvalds18 years
v2.6.23commit bbf25010f1...Linus Torvalds18 years
v2.6.23-rc9commit 3146b39c18...Linus Torvalds18 years
v2.6.23-rc8commit 4942de4a0e...Linus Torvalds18 years
v2.6.23-rc7commit 81cfe79b9c...Linus Torvalds18 years
v2.6.23-rc6commit 0d4cbb5e7f...Linus Torvalds18 years
v2.6.23-rc5commit 40ffbfad6b...Linus Torvalds18 years
v2.6.23-rc4commit b07d68b5ca...Linus Torvalds18 years
v2.6.23-rc3commit 39d3520c92...Linus Torvalds18 years
v2.6.23-rc2commit d4ac2477fa...Linus Torvalds18 years
v2.6.23-rc1commit f695baf2df...Linus Torvalds18 years
v2.6.22commit 7dcca30a32...Linus Torvalds18 years
v2.6.22-rc7commit a38d6181ff...Linus Torvalds18 years
v2.6.22-rc6commit 189548642c...Linus Torvalds18 years
v2.6.22-rc5commit 188e1f81ba...Linus Torvalds18 years
v2.6.22-rc4commit 5ecd3100e6...Linus Torvalds18 years
v2.6.22-rc3commit c420bc9f09...Linus Torvalds18 years
v2.6.22-rc2commit 55b637c6a0...Linus Torvalds18 years
v2.6.22-rc1commit 39403865d2...Linus Torvalds18 years
v2.6.21commit de46c33745...Linus Torvalds18 years
v2.6.21-rc7commit 94a05509a9...Linus Torvalds18 years
v2.6.21-rc6commit a21bd69e15...Linus Torvalds18 years
v2.6.21-rc5commit e0f2e3a06b...Linus Torvalds18 years
v2.6.21-rc4commit db98e0b434...Linus Torvalds18 years
v2.6.21-rc3commit 08e15e81a4...Linus Torvalds19 years
v2.6.21-rc2commit 606135a308...Linus Torvalds19 years
v2.6.21-rc1commit c8f71b01a5...Linus Torvalds19 years
v2.6.20commit 62d0cfcb27...Linus Torvalds19 years
v2.6.20-rc7commit f56df2f4db...Linus Torvalds19 years
v2.6.20-rc6commit 99abfeafb5...Linus Torvalds19 years
v2.6.20-rc5commit a8b3485287...Linus Torvalds19 years
v2.6.20-rc4commit bf81b46482...Linus Torvalds19 years
v2.6.20-rc3commit 669df1b478...Linus Torvalds19 years
v2.6.20-rc2commit 3bf8ba38f3...Linus Torvalds19 years
v2.6.20-rc1commit cc016448b0...Linus Torvalds19 years
v2.6.19commit 0215ffb08c...Linus Torvalds19 years
v2.6.19-rc6commit 44597f65f6...Linus Torvalds19 years
v2.6.19-rc5commit 80c2188127...Linus Torvalds19 years
v2.6.19-rc4commit ae99a78af3...Linus Torvalds19 years
v2.6.19-rc3commit 7059abedd2...Linus Torvalds19 years
v2.6.19-rc2commit b4bd8c6643...Linus Torvalds19 years
v2.6.19-rc1commit d223a60106...Linus Torvalds19 years
v2.6.18commit e478bec0ba...Linus Torvalds19 years
v2.6.18-rc7commit 95064a75eb...Linus Torvalds19 years
v2.6.18-rc6commit c336923b66...Linus Torvalds19 years
v2.6.18-rc5commit 60d4684068...Linus Torvalds19 years
v2.6.18-rc4commit 9f737633e6...Linus Torvalds19 years
v2.6.18-rc3commit b6ff50833a...Linus Torvalds19 years
v2.6.18-rc2commit 82d6897fef...Linus Torvalds19 years
v2.6.18-rc1commit 120bda20c6...Linus Torvalds19 years
v2.6.17commit 427abfa28a...Linus Torvalds19 years
v2.6.17-rc6commit 1def630a6a...Linus Torvalds19 years
v2.6.17-rc5commit a8bd60705a...Linus Torvalds19 years
v2.6.17-rc4commit d8c3291c73...Linus Torvalds19 years
v2.6.17-rc3commit 2be4d50295...Linus Torvalds19 years
v2.6.17-rc2commit 8bbde0e6d5...Linus Torvalds19 years
v2.6.17-rc1commit 6246b6128b...Linus Torvalds19 years
v2.6.16commit 7705a8792b...Linus Torvalds19 years
v2.6.16-rc6commit 535744878e...Linus Torvalds20 years
v2.6.16-rc5commit b9a33cebac...Linus Torvalds20 years
v2.6.16-rc4commit bd71c2b174...Linus Torvalds20 years
v2.6.16-rc3commit e9bb4c9929...Linus Torvalds20 years
v2.6.16-rc2commit 826eeb53a6...Linus Torvalds20 years
v2.6.16-rc1commit 2664b25051...Linus Torvalds20 years
v2.6.15commit 88026842b0...Linus Torvalds20 years
v2.6.15-rc7commit f89f5948fc...Linus Torvalds20 years
v2.6.15-rc6commit df7addbb45...Linus Torvalds20 years
v2.6.15-rc5commit 436b0f76f2...Linus Torvalds20 years
v2.6.15-rc4commit 5666c0947e...Linus Torvalds20 years
v2.6.15-rc3commit 624f54be20...Linus Torvalds20 years
v2.6.15-rc2commit 3bedff1d73...Linus Torvalds20 years
v2.6.15-rc1commit cd52d1ee9a...Linus Torvalds20 years
v2.6.14commit 741b2252a5...Linus Torvalds20 years
v2.6.14-rc5commit 93918e9afc...Linus Torvalds20 years
v2.6.14-rc4commit 907a426179...Linus Torvalds20 years
v2.6.14-rc3commit 1c9426e8a5...Linus Torvalds20 years
v2.6.14-rc2commit 676d55ae30...Linus Torvalds20 years
v2.6.14-rc1commit 2f4ba45a75...Linus Torvalds20 years
v2.6.13commit 02b3e4e2d7...Linus Torvalds20 years
v2.6.13-rc7commit 0572e3da3f...Linus Torvalds20 years
v2.6.13-rc6commit 6fc32179de...Linus Torvalds20 years
v2.6.13-rc5commit 9a351e30d7...Linus Torvalds20 years
v2.6.13-rc4commit 6395352334...Linus Torvalds20 years
v2.6.11tree c39ae07f39...
v2.6.11-treetree c39ae07f39...
v2.6.12commit 9ee1c939d1...
v2.6.12-rc2commit 1da177e4c3...
v2.6.12-rc3commit a2755a80f4...
v2.6.12-rc4commit 88d7bd8cb9...
v2.6.12-rc5commit 2a24ab628a...
v2.6.12-rc6commit 7cef5677ef...
v2.6.13-rc1commit 4c91aedb75...
v2.6.13-rc2commit a18bcb7450...
v2.6.13-rc3commit c32511e271...
generated by cgit v1.2.2 (git 2.25.0) at 2025-09-05 16:06:45 -0400
n) * Minshall version sounds: there are no _small_ * segments in flight. (tcp_nagle_check) * c) We have too many packets 'in flight' * * Don't use the nagle rule for urgent data (or * for the final FIN -DaveM). * * Also, Nagle rule does not apply to frames, which * sit in the middle of queue (they have no chances * to get new data) and if room at tail of skb is * not enough to save something seriously (<32 for now). */ /* Don't be strict about the congestion window for the * final FIN frame. -DaveM */ return (((nonagle&TCP_NAGLE_PUSH) || tp->urg_mode || !tcp_nagle_check(tp, skb, cur_mss, nonagle)) && (((tcp_packets_in_flight(tp) + (pkts-1)) < tp->snd_cwnd) || (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)) && !after(TCP_SKB_CB(skb)->end_seq, tp->snd_una + tp->snd_wnd)); } static inline int tcp_skb_is_last(const struct sock *sk, const struct sk_buff *skb) { return skb->next == (struct sk_buff *)&sk->sk_write_queue; } int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp) { struct sk_buff *skb = sk->sk_send_head; return (skb && tcp_snd_test(sk, skb, tcp_current_mss(sk, 1), (tcp_skb_is_last(sk, skb) ? TCP_NAGLE_PUSH : tp->nonagle))); } /* Send _single_ skb sitting at the send head. This function requires * true push pending frames to setup probe timer etc. */ void tcp_push_one(struct sock *sk, unsigned cur_mss) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb = sk->sk_send_head; if (tcp_snd_test(sk, skb, cur_mss, TCP_NAGLE_PUSH)) { /* Send it out now. */ TCP_SKB_CB(skb)->when = tcp_time_stamp; tcp_tso_set_push(skb); if (!tcp_transmit_skb(sk, skb_clone(skb, sk->sk_allocation))) { sk->sk_send_head = NULL; tp->snd_nxt = TCP_SKB_CB(skb)->end_seq; tcp_packets_out_inc(sk, tp, skb); return; } } } /* Function to create two new TCP segments. Shrinks the given segment * to the specified size and appends a new segment with the rest of the * packet to the list. This won't be called frequently, I hope. * Remember, these are still headerless SKBs at this point. */ static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *buff; int nsize; u16 flags; nsize = skb_headlen(skb) - len; if (nsize < 0) nsize = 0; if (skb_cloned(skb) && skb_is_nonlinear(skb) && pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) return -ENOMEM; /* Get a new skb... force flag on. */ buff = sk_stream_alloc_skb(sk, nsize, GFP_ATOMIC); if (buff == NULL) return -ENOMEM; /* We'll just try again later. */ sk_charge_skb(sk, buff); /* Correct the sequence numbers. */ TCP_SKB_CB(buff)->seq = TCP_SKB_CB(skb)->seq + len; TCP_SKB_CB(buff)->end_seq = TCP_SKB_CB(skb)->end_seq; TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(buff)->seq; /* PSH and FIN should only be set in the second packet. */ flags = TCP_SKB_CB(skb)->flags; TCP_SKB_CB(skb)->flags = flags & ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH); TCP_SKB_CB(buff)->flags = flags; TCP_SKB_CB(buff)->sacked = (TCP_SKB_CB(skb)->sacked & (TCPCB_LOST | TCPCB_EVER_RETRANS | TCPCB_AT_TAIL)); TCP_SKB_CB(skb)->sacked &= ~TCPCB_AT_TAIL; if (!skb_shinfo(skb)->nr_frags && skb->ip_summed != CHECKSUM_HW) { /* Copy and checksum data tail into the new buffer. */ buff->csum = csum_partial_copy_nocheck(skb->data + len, skb_put(buff, nsize), nsize, 0); skb_trim(skb, len); skb->csum = csum_block_sub(skb->csum, buff->csum, len); } else { skb->ip_summed = CHECKSUM_HW; skb_split(skb, buff, len); } buff->ip_summed = skb->ip_summed; /* Looks stupid, but our code really uses when of * skbs, which it never sent before. --ANK */ TCP_SKB_CB(buff)->when = TCP_SKB_CB(skb)->when; buff->stamp = skb->stamp; if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) { tp->lost_out -= tcp_skb_pcount(skb); tp->left_out -= tcp_skb_pcount(skb); } /* Fix up tso_factor for both original and new SKB. */ tcp_set_skb_tso_segs(sk, skb); tcp_set_skb_tso_segs(sk, buff); if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) { tp->lost_out += tcp_skb_pcount(skb); tp->left_out += tcp_skb_pcount(skb); } if (TCP_SKB_CB(buff)->sacked&TCPCB_LOST) { tp->lost_out += tcp_skb_pcount(buff); tp->left_out += tcp_skb_pcount(buff); } /* Link BUFF into the send queue. */ skb_header_release(buff); __skb_append(skb, buff); return 0; } /* This is similar to __pskb_pull_head() (it will go to core/skbuff.c * eventually). The difference is that pulled data not copied, but * immediately discarded. */ static unsigned char *__pskb_trim_head(struct sk_buff *skb, int len) { int i, k, eat; eat = len; k = 0; for (i=0; i<skb_shinfo(skb)->nr_frags; i++) { if (skb_shinfo(skb)->frags[i].size <= eat) { put_page(skb_shinfo(skb)->frags[i].page); eat -= skb_shinfo(skb)->frags[i].size; } else { skb_shinfo(skb)->frags[k] = skb_shinfo(skb)->frags[i]; if (eat) { skb_shinfo(skb)->frags[k].page_offset += eat; skb_shinfo(skb)->frags[k].size -= eat; eat = 0; } k++; } } skb_shinfo(skb)->nr_frags = k; skb->tail = skb->data; skb->data_len -= len; skb->len = skb->data_len; return skb->tail; } int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len) { if (skb_cloned(skb) && pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) return -ENOMEM; if (len <= skb_headlen(skb)) { __skb_pull(skb, len); } else { if (__pskb_trim_head(skb, len-skb_headlen(skb)) == NULL) return -ENOMEM; } TCP_SKB_CB(skb)->seq += len; skb->ip_summed = CHECKSUM_HW; skb->truesize -= len; sk->sk_wmem_queued -= len; sk->sk_forward_alloc += len; sock_set_flag(sk, SOCK_QUEUE_SHRUNK); /* Any change of skb->len requires recalculation of tso * factor and mss. */ if (tcp_skb_pcount(skb) > 1) tcp_set_skb_tso_segs(sk, skb); return 0; } /* This function synchronize snd mss to current pmtu/exthdr set. tp->rx_opt.user_mss is mss set by user by TCP_MAXSEG. It does NOT counts for TCP options, but includes only bare TCP header. tp->rx_opt.mss_clamp is mss negotiated at connection setup. It is minumum of user_mss and mss received with SYN. It also does not include TCP options. tp->pmtu_cookie is last pmtu, seen by this function. tp->mss_cache is current effective sending mss, including all tcp options except for SACKs. It is evaluated, taking into account current pmtu, but never exceeds tp->rx_opt.mss_clamp. NOTE1. rfc1122 clearly states that advertised MSS DOES NOT include either tcp or ip options. NOTE2. tp->pmtu_cookie and tp->mss_cache are READ ONLY outside this function. --ANK (980731) */ unsigned int tcp_sync_mss(struct sock *sk, u32 pmtu) { struct tcp_sock *tp = tcp_sk(sk); int mss_now; /* Calculate base mss without TCP options: It is MMS_S - sizeof(tcphdr) of rfc1122 */ mss_now = pmtu - tp->af_specific->net_header_len - sizeof(struct tcphdr); /* Clamp it (mss_clamp does not include tcp options) */ if (mss_now > tp->rx_opt.mss_clamp) mss_now = tp->rx_opt.mss_clamp; /* Now subtract optional transport overhead */ mss_now -= tp->ext_header_len; /* Then reserve room for full set of TCP options and 8 bytes of data */ if (mss_now < 48) mss_now = 48; /* Now subtract TCP options size, not including SACKs */ mss_now -= tp->tcp_header_len - sizeof(struct tcphdr); /* Bound mss with half of window */ if (tp->max_window && mss_now > (tp->max_window>>1)) mss_now = max((tp->max_window>>1), 68U - tp->tcp_header_len); /* And store cached results */ tp->pmtu_cookie = pmtu; tp->mss_cache = tp->mss_cache_std = mss_now; return mss_now; } /* Compute the current effective MSS, taking SACKs and IP options, * and even PMTU discovery events into account. * * LARGESEND note: !urg_mode is overkill, only frames up to snd_up * cannot be large. However, taking into account rare use of URG, this * is not a big flaw. */ unsigned int tcp_current_mss(struct sock *sk, int large) { struct tcp_sock *tp = tcp_sk(sk); struct dst_entry *dst = __sk_dst_get(sk); unsigned int do_large, mss_now; mss_now = tp->mss_cache_std; if (dst) { u32 mtu = dst_mtu(dst); if (mtu != tp->pmtu_cookie) mss_now = tcp_sync_mss(sk, mtu); } do_large = (large && (sk->sk_route_caps & NETIF_F_TSO) && !tp->urg_mode); if (do_large) { unsigned int large_mss, factor, limit; large_mss = 65535 - tp->af_specific->net_header_len - tp->ext_header_len - tp->tcp_header_len; if (tp->max_window && large_mss > (tp->max_window>>1)) large_mss = max((tp->max_window>>1), 68U - tp->tcp_header_len); factor = large_mss / mss_now; /* Always keep large mss multiple of real mss, but * do not exceed 1/tso_win_divisor of the congestion window * so we can keep the ACK clock ticking and minimize * bursting. */ limit = tp->snd_cwnd; if (sysctl_tcp_tso_win_divisor) limit /= sysctl_tcp_tso_win_divisor; limit = max(1U, limit); if (factor > limit) factor = limit; tp->mss_cache = mss_now * factor; mss_now = tp->mss_cache; } if (tp->rx_opt.eff_sacks) mss_now -= (TCPOLEN_SACK_BASE_ALIGNED + (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK)); return mss_now; } /* Congestion window validation. (RFC2861) */ static inline void tcp_cwnd_validate(struct sock *sk, struct tcp_sock *tp) { __u32 packets_out = tp->packets_out; if (packets_out >= tp->snd_cwnd) { /* Network is feed fully. */ tp->snd_cwnd_used = 0; tp->snd_cwnd_stamp = tcp_time_stamp; } else { /* Network starves. */ if (tp->packets_out > tp->snd_cwnd_used) tp->snd_cwnd_used = tp->packets_out; if ((s32)(tcp_time_stamp - tp->snd_cwnd_stamp) >= tp->rto) tcp_cwnd_application_limited(sk); } } /* This routine writes packets to the network. It advances the * send_head. This happens as incoming acks open up the remote * window for us. * * Returns 1, if no segments are in flight and we have queued segments, but * cannot send anything now because of SWS or another problem. */ static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; int sent_pkts; /* If we are closed, the bytes will have to remain here. * In time closedown will finish, we empty the write queue and all * will be happy. */ if (unlikely(sk->sk_state == TCP_CLOSE)) return 0; sent_pkts = 0; while ((skb = sk->sk_send_head) && tcp_snd_test(sk, skb, mss_now, tcp_skb_is_last(sk, skb) ? nonagle : TCP_NAGLE_PUSH)) { if (skb->len > mss_now) { if (tcp_fragment(sk, skb, mss_now)) break; } TCP_SKB_CB(skb)->when = tcp_time_stamp; tcp_tso_set_push(skb); if (tcp_transmit_skb(sk, skb_clone(skb, GFP_ATOMIC))) break; /* Advance the send_head. This one is sent out. * This call will increment packets_out. */ update_send_head(sk, tp, skb); tcp_minshall_update(tp, mss_now, skb); sent_pkts = 1; } if (sent_pkts) { tcp_cwnd_validate(sk, tp); return 0; } return !tp->packets_out && sk->sk_send_head; } /* Push out any pending frames which were held back due to * TCP_CORK or attempt at coalescing tiny packets. * The socket must be locked by the caller. */ void __tcp_push_pending_frames(struct sock *sk, struct tcp_sock *tp, unsigned int cur_mss, int nonagle) { struct sk_buff *skb = sk->sk_send_head; if (skb) { if (tcp_write_xmit(sk, cur_mss, nonagle)) tcp_check_probe_timer(sk, tp); } } /* This function returns the amount that we can raise the * usable window based on the following constraints * * 1. The window can never be shrunk once it is offered (RFC 793) * 2. We limit memory per socket * * RFC 1122: * "the suggested [SWS] avoidance algorithm for the receiver is to keep * RECV.NEXT + RCV.WIN fixed until: * RCV.BUFF - RCV.USER - RCV.WINDOW >= min(1/2 RCV.BUFF, MSS)" * * i.e. don't raise the right edge of the window until you can raise * it at least MSS bytes. * * Unfortunately, the recommended algorithm breaks header prediction, * since header prediction assumes th->window stays fixed. * * Strictly speaking, keeping th->window fixed violates the receiver * side SWS prevention criteria. The problem is that under this rule * a stream of single byte packets will cause the right side of the * window to always advance by a single byte. * * Of course, if the sender implements sender side SWS prevention * then this will not be a problem. * * BSD seems to make the following compromise: * * If the free space is less than the 1/4 of the maximum * space available and the free space is less than 1/2 mss, * then set the window to 0. * [ Actually, bsd uses MSS and 1/4 of maximal _window_ ] * Otherwise, just prevent the window from shrinking * and from being larger than the largest representable value. * * This prevents incremental opening of the window in the regime * where TCP is limited by the speed of the reader side taking * data out of the TCP receive queue. It does nothing about * those cases where the window is constrained on the sender side * because the pipeline is full. * * BSD also seems to "accidentally" limit itself to windows that are a * multiple of MSS, at least until the free space gets quite small. * This would appear to be a side effect of the mbuf implementation. * Combining these two algorithms results in the observed behavior * of having a fixed window size at almost all times. * * Below we obtain similar behavior by forcing the offered window to * a multiple of the mss when it is feasible to do so. * * Note, we don't "adjust" for TIMESTAMP or SACK option bytes. * Regular options like TIMESTAMP are taken into account. */ u32 __tcp_select_window(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); /* MSS for the peer's data. Previous verions used mss_clamp * here. I don't know if the value based on our guesses * of peer's MSS is better for the performance. It's more correct * but may be worse for the performance because of rcv_mss * fluctuations. --SAW 1998/11/1 */ int mss = tp->ack.rcv_mss; int free_space = tcp_space(sk); int full_space = min_t(int, tp->window_clamp, tcp_full_space(sk)); int window; if (mss > full_space) mss = full_space; if (free_space < full_space/2) { tp->ack.quick = 0; if (tcp_memory_pressure) tp->rcv_ssthresh = min(tp->rcv_ssthresh, 4U*tp->advmss); if (free_space < mss) return 0; } if (free_space > tp->rcv_ssthresh) free_space = tp->rcv_ssthresh; /* Don't do rounding if we are using window scaling, since the * scaled window will not line up with the MSS boundary anyway. */ window = tp->rcv_wnd; if (tp->rx_opt.rcv_wscale) { window = free_space; /* Advertise enough space so that it won't get scaled away. * Import case: prevent zero window announcement if * 1<<rcv_wscale > mss. */ if (((window >> tp->rx_opt.rcv_wscale) << tp->rx_opt.rcv_wscale) != window) window = (((window >> tp->rx_opt.rcv_wscale) + 1) << tp->rx_opt.rcv_wscale); } else { /* Get the largest window that is a nice multiple of mss. * Window clamp already applied above. * If our current window offering is within 1 mss of the * free space we just keep it. This prevents the divide * and multiply from happening most of the time. * We also don't do any window rounding when the free space * is too small. */ if (window <= free_space - mss || window > free_space) window = (free_space/mss)*mss; } return window; } /* Attempt to collapse two adjacent SKB's during retransmission. */ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int mss_now) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *next_skb = skb->next; /* The first test we must make is that neither of these two * SKB's are still referenced by someone else. */ if (!skb_cloned(skb) && !skb_cloned(next_skb)) { int skb_size = skb->len, next_skb_size = next_skb->len; u16 flags = TCP_SKB_CB(skb)->flags; /* Also punt if next skb has been SACK'd. */ if(TCP_SKB_CB(next_skb)->sacked & TCPCB_SACKED_ACKED) return; /* Next skb is out of window. */ if (after(TCP_SKB_CB(next_skb)->end_seq, tp->snd_una+tp->snd_wnd)) return; /* Punt if not enough space exists in the first SKB for * the data in the second, or the total combined payload * would exceed the MSS. */ if ((next_skb_size > skb_tailroom(skb)) || ((skb_size + next_skb_size) > mss_now)) return; BUG_ON(tcp_skb_pcount(skb) != 1 || tcp_skb_pcount(next_skb) != 1); /* Ok. We will be able to collapse the packet. */ __skb_unlink(next_skb, next_skb->list); memcpy(skb_put(skb, next_skb_size), next_skb->data, next_skb_size); if (next_skb->ip_summed == CHECKSUM_HW) skb->ip_summed = CHECKSUM_HW; if (skb->ip_summed != CHECKSUM_HW) skb->csum = csum_block_add(skb->csum, next_skb->csum, skb_size); /* Update sequence range on original skb. */ TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(next_skb)->end_seq; /* Merge over control information. */ flags |= TCP_SKB_CB(next_skb)->flags; /* This moves PSH/FIN etc. over */ TCP_SKB_CB(skb)->flags = flags; /* All done, get rid of second SKB and account for it so * packet counting does not break. */ TCP_SKB_CB(skb)->sacked |= TCP_SKB_CB(next_skb)->sacked&(TCPCB_EVER_RETRANS|TCPCB_AT_TAIL); if (TCP_SKB_CB(next_skb)->sacked&TCPCB_SACKED_RETRANS) tp->retrans_out -= tcp_skb_pcount(next_skb); if (TCP_SKB_CB(next_skb)->sacked&TCPCB_LOST) { tp->lost_out -= tcp_skb_pcount(next_skb); tp->left_out -= tcp_skb_pcount(next_skb); } /* Reno case is special. Sigh... */ if (!tp->rx_opt.sack_ok && tp->sacked_out) { tcp_dec_pcount_approx(&tp->sacked_out, next_skb); tp->left_out -= tcp_skb_pcount(next_skb); } /* Not quite right: it can be > snd.fack, but * it is better to underestimate fackets. */ tcp_dec_pcount_approx(&tp->fackets_out, next_skb); tcp_packets_out_dec(tp, next_skb); sk_stream_free_skb(sk, next_skb); } } /* Do a simple retransmit without using the backoff mechanisms in * tcp_timer. This is used for path mtu discovery. * The socket is already locked here. */ void tcp_simple_retransmit(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; unsigned int mss = tcp_current_mss(sk, 0); int lost = 0; sk_stream_for_retrans_queue(skb, sk) { if (skb->len > mss && !(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED)) { if (TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS) { TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; tp->retrans_out -= tcp_skb_pcount(skb); } if (!(TCP_SKB_CB(skb)->sacked&TCPCB_LOST)) { TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; tp->lost_out += tcp_skb_pcount(skb); lost = 1; } } } if (!lost) return; tcp_sync_left_out(tp); /* Don't muck with the congestion window here. * Reason is that we do not increase amount of _data_ * in network, but units changed and effective * cwnd/ssthresh really reduced now. */ if (tp->ca_state != TCP_CA_Loss) { tp->high_seq = tp->snd_nxt; tp->snd_ssthresh = tcp_current_ssthresh(tp); tp->prior_ssthresh = 0; tp->undo_marker = 0; tcp_set_ca_state(tp, TCP_CA_Loss); } tcp_xmit_retransmit_queue(sk); } /* This retransmits one SKB. Policy decisions and retransmit queue * state updates are done by the caller. Returns non-zero if an * error occurred which prevented the send. */ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); unsigned int cur_mss = tcp_current_mss(sk, 0); int err; /* Do not sent more than we queued. 1/4 is reserved for possible * copying overhead: frgagmentation, tunneling, mangling etc. */ if (atomic_read(&sk->sk_wmem_alloc) > min(sk->sk_wmem_queued + (sk->sk_wmem_queued >> 2), sk->sk_sndbuf)) return -EAGAIN; if (before(TCP_SKB_CB(skb)->seq, tp->snd_una)) { if (before(TCP_SKB_CB(skb)->end_seq, tp->snd_una)) BUG(); if (sk->sk_route_caps & NETIF_F_TSO) { sk->sk_route_caps &= ~NETIF_F_TSO; sock_set_flag(sk, SOCK_NO_LARGESEND); tp->mss_cache = tp->mss_cache_std; } if (tcp_trim_head(sk, skb, tp->snd_una - TCP_SKB_CB(skb)->seq)) return -ENOMEM; } /* If receiver has shrunk his window, and skb is out of * new window, do not retransmit it. The exception is the * case, when window is shrunk to zero. In this case * our retransmit serves as a zero window probe. */ if (!before(TCP_SKB_CB(skb)->seq, tp->snd_una+tp->snd_wnd) && TCP_SKB_CB(skb)->seq != tp->snd_una) return -EAGAIN; if (skb->len > cur_mss) { int old_factor = tcp_skb_pcount(skb); int new_factor; if (tcp_fragment(sk, skb, cur_mss)) return -ENOMEM; /* We'll try again later. */ /* New SKB created, account for it. */ new_factor = tcp_skb_pcount(skb); tp->packets_out -= old_factor - new_factor; tp->packets_out += tcp_skb_pcount(skb->next); } /* Collapse two adjacent packets if worthwhile and we can. */ if(!(TCP_SKB_CB(skb)->flags & TCPCB_FLAG_SYN) && (skb->len < (cur_mss >> 1)) && (skb->next != sk->sk_send_head) && (skb->next != (struct sk_buff *)&sk->sk_write_queue) && (skb_shinfo(skb)->nr_frags == 0 && skb_shinfo(skb->next)->nr_frags == 0) && (tcp_skb_pcount(skb) == 1 && tcp_skb_pcount(skb->next) == 1) && (sysctl_tcp_retrans_collapse != 0)) tcp_retrans_try_collapse(sk, skb, cur_mss); if(tp->af_specific->rebuild_header(sk)) return -EHOSTUNREACH; /* Routing failure or similar. */ /* Some Solaris stacks overoptimize and ignore the FIN on a * retransmit when old data is attached. So strip it off * since it is cheap to do so and saves bytes on the network. */ if(skb->len > 0 && (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) && tp->snd_una == (TCP_SKB_CB(skb)->end_seq - 1)) { if (!pskb_trim(skb, 0)) { TCP_SKB_CB(skb)->seq = TCP_SKB_CB(skb)->end_seq - 1; skb_shinfo(skb)->tso_segs = 1; skb_shinfo(skb)->tso_size = 0; skb->ip_summed = CHECKSUM_NONE; skb->csum = 0; } } /* Make a copy, if the first transmission SKB clone we made * is still in somebody's hands, else make a clone. */ TCP_SKB_CB(skb)->when = tcp_time_stamp; tcp_tso_set_push(skb); err = tcp_transmit_skb(sk, (skb_cloned(skb) ? pskb_copy(skb, GFP_ATOMIC): skb_clone(skb, GFP_ATOMIC))); if (err == 0) { /* Update global TCP statistics. */ TCP_INC_STATS(TCP_MIB_RETRANSSEGS); tp->total_retrans++; #if FASTRETRANS_DEBUG > 0 if (TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS) { if (net_ratelimit()) printk(KERN_DEBUG "retrans_out leaked.\n"); } #endif TCP_SKB_CB(skb)->sacked |= TCPCB_RETRANS; tp->retrans_out += tcp_skb_pcount(skb); /* Save stamp of the first retransmit. */ if (!tp->retrans_stamp) tp->retrans_stamp = TCP_SKB_CB(skb)->when; tp->undo_retrans++; /* snd_nxt is stored to detect loss of retransmitted segment, * see tcp_input.c tcp_sacktag_write_queue(). */ TCP_SKB_CB(skb)->ack_seq = tp->snd_nxt; } return err; } /* This gets called after a retransmit timeout, and the initially * retransmitted data is acknowledged. It tries to continue * resending the rest of the retransmit queue, until either * we've sent it all or the congestion window limit is reached. * If doing SACK, the first ACK which comes back for a timeout * based retransmit packet might feed us FACK information again. * If so, we use it to avoid unnecessarily retransmissions. */ void tcp_xmit_retransmit_queue(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; int packet_cnt = tp->lost_out; /* First pass: retransmit lost packets. */ if (packet_cnt) { sk_stream_for_retrans_queue(skb, sk) { __u8 sacked = TCP_SKB_CB(skb)->sacked; /* Assume this retransmit will generate * only one packet for congestion window * calculation purposes. This works because * tcp_retransmit_skb() will chop up the * packet to be MSS sized and all the * packet counting works out. */ if (tcp_packets_in_flight(tp) >= tp->snd_cwnd) return; if (sacked&TCPCB_LOST) { if (!(sacked&(TCPCB_SACKED_ACKED|TCPCB_SACKED_RETRANS))) { if (tcp_retransmit_skb(sk, skb)) return; if (tp->ca_state != TCP_CA_Loss) NET_INC_STATS_BH(LINUX_MIB_TCPFASTRETRANS); else NET_INC_STATS_BH(LINUX_MIB_TCPSLOWSTARTRETRANS); if (skb == skb_peek(&sk->sk_write_queue)) tcp_reset_xmit_timer(sk, TCP_TIME_RETRANS, tp->rto); } packet_cnt -= tcp_skb_pcount(skb); if (packet_cnt <= 0) break; } } } /* OK, demanded retransmission is finished. */ /* Forward retransmissions are possible only during Recovery. */ if (tp->ca_state != TCP_CA_Recovery) return; /* No forward retransmissions in Reno are possible. */ if (!tp->rx_opt.sack_ok) return; /* Yeah, we have to make difficult choice between forward transmission * and retransmission... Both ways have their merits... * * For now we do not retransmit anything, while we have some new * segments to send. */ if (tcp_may_send_now(sk, tp)) return; packet_cnt = 0; sk_stream_for_retrans_queue(skb, sk) { /* Similar to the retransmit loop above we * can pretend that the retransmitted SKB * we send out here will be composed of one * real MSS sized packet because tcp_retransmit_skb() * will fragment it if necessary. */ if (++packet_cnt > tp->fackets_out) break; if (tcp_packets_in_flight(tp) >= tp->snd_cwnd) break; if (TCP_SKB_CB(skb)->sacked & TCPCB_TAGBITS) continue; /* Ok, retransmit it. */ if (tcp_retransmit_skb(sk, skb)) break; if (skb == skb_peek(&sk->sk_write_queue)) tcp_reset_xmit_timer(sk, TCP_TIME_RETRANS, tp->rto); NET_INC_STATS_BH(LINUX_MIB_TCPFORWARDRETRANS); } } /* Send a fin. The caller locks the socket for us. This cannot be * allowed to fail queueing a FIN frame under any circumstances. */ void tcp_send_fin(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb = skb_peek_tail(&sk->sk_write_queue); int mss_now; /* Optimization, tack on the FIN if we have a queue of * unsent frames. But be careful about outgoing SACKS * and IP options. */ mss_now = tcp_current_mss(sk, 1); if (sk->sk_send_head != NULL) { TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_FIN; TCP_SKB_CB(skb)->end_seq++; tp->write_seq++; } else { /* Socket is locked, keep trying until memory is available. */ for (;;) { skb = alloc_skb(MAX_TCP_HEADER, GFP_KERNEL); if (skb) break; yield(); } /* Reserve space for headers and prepare control bits. */ skb_reserve(skb, MAX_TCP_HEADER); skb->csum = 0; TCP_SKB_CB(skb)->flags = (TCPCB_FLAG_ACK | TCPCB_FLAG_FIN); TCP_SKB_CB(skb)->sacked = 0; skb_shinfo(skb)->tso_segs = 1; skb_shinfo(skb)->tso_size = 0; /* FIN eats a sequence byte, write_seq advanced by tcp_queue_skb(). */ TCP_SKB_CB(skb)->seq = tp->write_seq; TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + 1; tcp_queue_skb(sk, skb); } __tcp_push_pending_frames(sk, tp, mss_now, TCP_NAGLE_OFF); } /* We get here when a process closes a file descriptor (either due to * an explicit close() or as a byproduct of exit()'ing) and there * was unread data in the receive queue. This behavior is recommended * by draft-ietf-tcpimpl-prob-03.txt section 3.10. -DaveM */ void tcp_send_active_reset(struct sock *sk, int priority) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; /* NOTE: No TCP options attached and we never retransmit this. */ skb = alloc_skb(MAX_TCP_HEADER, priority); if (!skb) { NET_INC_STATS(LINUX_MIB_TCPABORTFAILED); return; } /* Reserve space for headers and prepare control bits. */ skb_reserve(skb, MAX_TCP_HEADER); skb->csum = 0; TCP_SKB_CB(skb)->flags = (TCPCB_FLAG_ACK | TCPCB_FLAG_RST); TCP_SKB_CB(skb)->sacked = 0; skb_shinfo(skb)->tso_segs = 1; skb_shinfo(skb)->tso_size = 0; /* Send it off. */ TCP_SKB_CB(skb)->seq = tcp_acceptable_seq(sk, tp); TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq; TCP_SKB_CB(skb)->when = tcp_time_stamp; if (tcp_transmit_skb(sk, skb)) NET_INC_STATS(LINUX_MIB_TCPABORTFAILED); } /* WARNING: This routine must only be called when we have already sent * a SYN packet that crossed the incoming SYN that caused this routine * to get called. If this assumption fails then the initial rcv_wnd * and rcv_wscale values will not be correct. */ int tcp_send_synack(struct sock *sk) { struct sk_buff* skb; skb = skb_peek(&sk->sk_write_queue); if (skb == NULL || !(TCP_SKB_CB(skb)->flags&TCPCB_FLAG_SYN)) { printk(KERN_DEBUG "tcp_send_synack: wrong queue state\n"); return -EFAULT; } if (!(TCP_SKB_CB(skb)->flags&TCPCB_FLAG_ACK)) { if (skb_cloned(skb)) { struct sk_buff *nskb = skb_copy(skb, GFP_ATOMIC); if (nskb == NULL) return -ENOMEM; __skb_unlink(skb, &sk->sk_write_queue); skb_header_release(nskb); __skb_queue_head(&sk->sk_write_queue, nskb); sk_stream_free_skb(sk, skb); sk_charge_skb(sk, nskb); skb = nskb; } TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_ACK; TCP_ECN_send_synack(tcp_sk(sk), skb); } TCP_SKB_CB(skb)->when = tcp_time_stamp; return tcp_transmit_skb(sk, skb_clone(skb, GFP_ATOMIC)); } /* * Prepare a SYN-ACK. */ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, struct request_sock *req) { struct inet_request_sock *ireq = inet_rsk(req); struct tcp_sock *tp = tcp_sk(sk); struct tcphdr *th; int tcp_header_size; struct sk_buff *skb; skb = sock_wmalloc(sk, MAX_TCP_HEADER + 15, 1, GFP_ATOMIC); if (skb == NULL) return NULL; /* Reserve space for headers. */ skb_reserve(skb, MAX_TCP_HEADER); skb->dst = dst_clone(dst); tcp_header_size = (sizeof(struct tcphdr) + TCPOLEN_MSS + (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0) + (ireq->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + /* SACK_PERM is in the place of NOP NOP of TS */ ((ireq->sack_ok && !ireq->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); skb->h.th = th = (struct tcphdr *) skb_push(skb, tcp_header_size); memset(th, 0, sizeof(struct tcphdr)); th->syn = 1; th->ack = 1; if (dst->dev->features&NETIF_F_TSO) ireq->ecn_ok = 0; TCP_ECN_make_synack(req, th); th->source = inet_sk(sk)->sport; th->dest = ireq->rmt_port; TCP_SKB_CB(skb)->seq = tcp_rsk(req)->snt_isn; TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + 1; TCP_SKB_CB(skb)->sacked = 0; skb_shinfo(skb)->tso_segs = 1; skb_shinfo(skb)->tso_size = 0; th->seq = htonl(TCP_SKB_CB(skb)->seq); th->ack_seq = htonl(tcp_rsk(req)->rcv_isn + 1); if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */ __u8 rcv_wscale; /* Set this up on the first call only */ req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW); /* tcp_full_space because it is guaranteed to be the first packet */ tcp_select_initial_window(tcp_full_space(sk), dst_metric(dst, RTAX_ADVMSS) - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), &req->rcv_wnd, &req->window_clamp, ireq->wscale_ok, &rcv_wscale); ireq->rcv_wscale = rcv_wscale; } /* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */ th->window = htons(req->rcv_wnd); TCP_SKB_CB(skb)->when = tcp_time_stamp; tcp_syn_build_options((__u32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok, ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale, TCP_SKB_CB(skb)->when, req->ts_recent); skb->csum = 0; th->doff = (tcp_header_size >> 2); TCP_INC_STATS(TCP_MIB_OUTSEGS); return skb; } /* * Do all connect socket setups that can be done AF independent. */ static inline void tcp_connect_init(struct sock *sk) { struct dst_entry *dst = __sk_dst_get(sk); struct tcp_sock *tp = tcp_sk(sk); __u8 rcv_wscale; /* We'll fix this up when we get a response from the other end. * See tcp_input.c:tcp_rcv_state_process case TCP_SYN_SENT. */ tp->tcp_header_len = sizeof(struct tcphdr) + (sysctl_tcp_timestamps ? TCPOLEN_TSTAMP_ALIGNED : 0); /* If user gave his TCP_MAXSEG, record it to clamp */ if (tp->rx_opt.user_mss) tp->rx_opt.mss_clamp = tp->rx_opt.user_mss; tp->max_window = 0; tcp_sync_mss(sk, dst_mtu(dst)); if (!tp->window_clamp) tp->window_clamp = dst_metric(dst, RTAX_WINDOW); tp->advmss = dst_metric(dst, RTAX_ADVMSS); tcp_initialize_rcv_mss(sk); tcp_select_initial_window(tcp_full_space(sk), tp->advmss - (tp->rx_opt.ts_recent_stamp ? tp->tcp_header_len - sizeof(struct tcphdr) : 0), &tp->rcv_wnd, &tp->window_clamp, sysctl_tcp_window_scaling, &rcv_wscale); tp->rx_opt.rcv_wscale = rcv_wscale; tp->rcv_ssthresh = tp->rcv_wnd; sk->sk_err = 0; sock_reset_flag(sk, SOCK_DONE); tp->snd_wnd = 0; tcp_init_wl(tp, tp->write_seq, 0); tp->snd_una = tp->write_seq; tp->snd_sml = tp->write_seq; tp->rcv_nxt = 0; tp->rcv_wup = 0; tp->copied_seq = 0; tp->rto = TCP_TIMEOUT_INIT; tp->retransmits = 0; tcp_clear_retrans(tp); } /* * Build a SYN and send it off. */ int tcp_connect(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *buff; tcp_connect_init(sk); buff = alloc_skb(MAX_TCP_HEADER + 15, sk->sk_allocation); if (unlikely(buff == NULL)) return -ENOBUFS; /* Reserve space for headers. */ skb_reserve(buff, MAX_TCP_HEADER); TCP_SKB_CB(buff)->flags = TCPCB_FLAG_SYN; TCP_ECN_send_syn(sk, tp, buff); TCP_SKB_CB(buff)->sacked = 0; skb_shinfo(buff)->tso_segs = 1; skb_shinfo(buff)->tso_size = 0; buff->csum = 0; TCP_SKB_CB(buff)->seq = tp->write_seq++; TCP_SKB_CB(buff)->end_seq = tp->write_seq; tp->snd_nxt = tp->write_seq; tp->pushed_seq = tp->write_seq; /* Send it off. */ TCP_SKB_CB(buff)->when = tcp_time_stamp; tp->retrans_stamp = TCP_SKB_CB(buff)->when; skb_header_release(buff); __skb_queue_tail(&sk->sk_write_queue, buff); sk_charge_skb(sk, buff); tp->packets_out += tcp_skb_pcount(buff); tcp_transmit_skb(sk, skb_clone(buff, GFP_KERNEL)); TCP_INC_STATS(TCP_MIB_ACTIVEOPENS); /* Timer for repeating the SYN until an answer. */ tcp_reset_xmit_timer(sk, TCP_TIME_RETRANS, tp->rto); return 0; } /* Send out a delayed ack, the caller does the policy checking * to see if we should even be here. See tcp_input.c:tcp_ack_snd_check() * for details. */ void tcp_send_delayed_ack(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); int ato = tp->ack.ato; unsigned long timeout; if (ato > TCP_DELACK_MIN) { int max_ato = HZ/2; if (tp->ack.pingpong || (tp->ack.pending&TCP_ACK_PUSHED)) max_ato = TCP_DELACK_MAX; /* Slow path, intersegment interval is "high". */ /* If some rtt estimate is known, use it to bound delayed ack. * Do not use tp->rto here, use results of rtt measurements * directly. */ if (tp->srtt) { int rtt = max(tp->srtt>>3, TCP_DELACK_MIN); if (rtt < max_ato) max_ato = rtt; } ato = min(ato, max_ato); } /* Stay within the limit we were given */ timeout = jiffies + ato; /* Use new timeout only if there wasn't a older one earlier. */ if (tp->ack.pending&TCP_ACK_TIMER) { /* If delack timer was blocked or is about to expire, * send ACK now. */ if (tp->ack.blocked || time_before_eq(tp->ack.timeout, jiffies+(ato>>2))) { tcp_send_ack(sk); return; } if (!time_before(timeout, tp->ack.timeout)) timeout = tp->ack.timeout; } tp->ack.pending |= TCP_ACK_SCHED|TCP_ACK_TIMER; tp->ack.timeout = timeout; sk_reset_timer(sk, &tp->delack_timer, timeout); } /* This routine sends an ack and also updates the window. */ void tcp_send_ack(struct sock *sk) { /* If we have been reset, we may not send again. */ if (sk->sk_state != TCP_CLOSE) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *buff; /* We are not putting this on the write queue, so * tcp_transmit_skb() will set the ownership to this * sock. */ buff = alloc_skb(MAX_TCP_HEADER, GFP_ATOMIC); if (buff == NULL) { tcp_schedule_ack(tp); tp->ack.ato = TCP_ATO_MIN; tcp_reset_xmit_timer(sk, TCP_TIME_DACK, TCP_DELACK_MAX); return; } /* Reserve space for headers and prepare control bits. */ skb_reserve(buff, MAX_TCP_HEADER); buff->csum = 0; TCP_SKB_CB(buff)->flags = TCPCB_FLAG_ACK; TCP_SKB_CB(buff)->sacked = 0; skb_shinfo(buff)->tso_segs = 1; skb_shinfo(buff)->tso_size = 0; /* Send it off, this clears delayed acks for us. */ TCP_SKB_CB(buff)->seq = TCP_SKB_CB(buff)->end_seq = tcp_acceptable_seq(sk, tp); TCP_SKB_CB(buff)->when = tcp_time_stamp; tcp_transmit_skb(sk, buff); } } /* This routine sends a packet with an out of date sequence * number. It assumes the other end will try to ack it. * * Question: what should we make while urgent mode? * 4.4BSD forces sending single byte of data. We cannot send * out of window data, because we have SND.NXT==SND.MAX... * * Current solution: to send TWO zero-length segments in urgent mode: * one is with SEG.SEQ=SND.UNA to deliver urgent pointer, another is * out-of-date with SND.UNA-1 to probe window. */ static int tcp_xmit_probe_skb(struct sock *sk, int urgent) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; /* We don't queue it, tcp_transmit_skb() sets ownership. */ skb = alloc_skb(MAX_TCP_HEADER, GFP_ATOMIC); if (skb == NULL) return -1; /* Reserve space for headers and set control bits. */ skb_reserve(skb, MAX_TCP_HEADER); skb->csum = 0; TCP_SKB_CB(skb)->flags = TCPCB_FLAG_ACK; TCP_SKB_CB(skb)->sacked = urgent; skb_shinfo(skb)->tso_segs = 1; skb_shinfo(skb)->tso_size = 0; /* Use a previous sequence. This should cause the other * end to send an ack. Don't queue or clone SKB, just * send it. */ TCP_SKB_CB(skb)->seq = urgent ? tp->snd_una : tp->snd_una - 1; TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq; TCP_SKB_CB(skb)->when = tcp_time_stamp; return tcp_transmit_skb(sk, skb); } int tcp_write_wakeup(struct sock *sk) { if (sk->sk_state != TCP_CLOSE) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; if ((skb = sk->sk_send_head) != NULL && before(TCP_SKB_CB(skb)->seq, tp->snd_una+tp->snd_wnd)) { int err; unsigned int mss = tcp_current_mss(sk, 0); unsigned int seg_size = tp->snd_una+tp->snd_wnd-TCP_SKB_CB(skb)->seq; if (before(tp->pushed_seq, TCP_SKB_CB(skb)->end_seq)) tp->pushed_seq = TCP_SKB_CB(skb)->end_seq; /* We are probing the opening of a window * but the window size is != 0 * must have been a result SWS avoidance ( sender ) */ if (seg_size < TCP_SKB_CB(skb)->end_seq - TCP_SKB_CB(skb)->seq || skb->len > mss) { seg_size = min(seg_size, mss); TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH; if (tcp_fragment(sk, skb, seg_size)) return -1; /* SWS override triggered forced fragmentation. * Disable TSO, the connection is too sick. */ if (sk->sk_route_caps & NETIF_F_TSO) { sock_set_flag(sk, SOCK_NO_LARGESEND); sk->sk_route_caps &= ~NETIF_F_TSO; tp->mss_cache = tp->mss_cache_std; } } else if (!tcp_skb_pcount(skb)) tcp_set_skb_tso_segs(sk, skb); TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH; TCP_SKB_CB(skb)->when = tcp_time_stamp; tcp_tso_set_push(skb); err = tcp_transmit_skb(sk, skb_clone(skb, GFP_ATOMIC)); if (!err) { update_send_head(sk, tp, skb); } return err; } else { if (tp->urg_mode && between(tp->snd_up, tp->snd_una+1, tp->snd_una+0xFFFF)) tcp_xmit_probe_skb(sk, TCPCB_URG); return tcp_xmit_probe_skb(sk, 0); } } return -1; } /* A window probe timeout has occurred. If window is not closed send * a partial packet else a zero probe. */ void tcp_send_probe0(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); int err; err = tcp_write_wakeup(sk); if (tp->packets_out || !sk->sk_send_head) { /* Cancel probe timer, if it is not required. */ tp->probes_out = 0; tp->backoff = 0; return; } if (err <= 0) { if (tp->backoff < sysctl_tcp_retries2) tp->backoff++; tp->probes_out++; tcp_reset_xmit_timer (sk, TCP_TIME_PROBE0, min(tp->rto << tp->backoff, TCP_RTO_MAX)); } else { /* If packet was not sent due to local congestion, * do not backoff and do not remember probes_out. * Let local senders to fight for local resources. * * Use accumulated backoff yet. */ if (!tp->probes_out) tp->probes_out=1; tcp_reset_xmit_timer (sk, TCP_TIME_PROBE0, min(tp->rto << tp->backoff, TCP_RESOURCE_PROBE_INTERVAL)); } } EXPORT_SYMBOL(tcp_connect); EXPORT_SYMBOL(tcp_make_synack); EXPORT_SYMBOL(tcp_simple_retransmit); EXPORT_SYMBOL(tcp_sync_mss);
generated by cgit v1.2.2 (git 2.25.0) at 2025-09-05 16:06:45 -0400