)]}' { "log": [ { "commit": "89624e152475881f05bb88aca261239d1b45a833", "tree": "65fd02f4175c55d513e747fcd335f43d45a8593b", "parents": [ "af00b1e55147eecbfc3b339658f656fe320049a9" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Apr 01 16:00:32 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 23 11:14:32 2026 -0700" }, "message": "http2: move common server symbols into new files\n\nIn preparation for adding a build-tag-guarded version of the\nserver API which wraps net/http, move various parts of the\nexisting API into common files.\n\nThis moves exported symbols (so we don\u0027t need to duplicate\nthem and their documentation between the two implementations)\nand some functions which will be used by the wrapping\nimplementation.\n\nChange-Id: I6965dd1dabc2081d19af55b607694d246a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/769060\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "af00b1e55147eecbfc3b339658f656fe320049a9", "tree": "91c71504aeba77ac4e3b1002f90746c6c964c339", "parents": [ "f34f3751d867ac2ce19914dd12cedb77461ff2f4" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Apr 01 16:00:32 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 23 11:14:30 2026 -0700" }, "message": "http2: move common transport symbols into new files\n\nIn preparation for adding a build-tag-guarded version of the\ntransport/client API which wraps net/http, move various\nparts of the existing API into common files.\n\nThis moves exported symbols (so we don\u0027t need to duplicate\nthem and their documentation between the two implementations)\nand some functions which will be used by the wrapping\nimplementation.\n\nChange-Id: I58c2f48a37fd9cf749831721141d46a76a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/765400\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "f34f3751d867ac2ce19914dd12cedb77461ff2f4", "tree": "01e920dcbcf7c92fbde495b1f03a7c829bf3e45c", "parents": [ "0dd4bea8b74a97a7f6ad941e1ce598434226100e" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Fri Apr 03 15:30:56 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Wed Apr 22 15:59:51 2026 -0700" }, "message": "http2: add server API tests\n\nAdd a collection of tests which specifically exercise the user-facing\nserver APIs in this package.\n\nMany of these tests are somewhat redundant with existing tests in\nthe package. Their intent is to ensure consistency between various\nways users may access HTTP/2 features:\n\n - Using a x/net/http2.Server directly.\n - Using a net/http.Server configured with ConfigureServer.\n\nand:\n\n - Using the HTTP/2 implementation in x/net/http2.\n - Using the HTTP/2 implementation in net/http, wrapped by x/net/http2.\n (A case to be implemented in future CLs.)\n\nFor #78508\n\nChange-Id: I784cda6854e84c7305f3e6b99923b6bd6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/765381\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\n" }, { "commit": "0dd4bea8b74a97a7f6ad941e1ce598434226100e", "tree": "8ce19fcad391815fca57db670995f07a8ed5ba88", "parents": [ "7301a20d0d957b09475b0468d0d857e428480d8c" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Apr 22 13:37:31 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Wed Apr 22 15:05:02 2026 -0700" }, "message": "webdav: add security considerations to package doc\n\nDocument that this package is not hardened against malicious clients.\n\nChange-Id: I7fb8491e21feb7513edbd5a5bd751e486a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/769880\nReviewed-by: Neal Patel \u003cnealpatel@google.com\u003e\nReviewed-by: Roland Shoemaker \u003croland@golang.org\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "7301a20d0d957b09475b0468d0d857e428480d8c", "tree": "1596a2458e8872f1b80915a7490b04fcf5e8bab9", "parents": [ "f70faeaf29edebe946e858be8acfebd70276cf33" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Apr 09 17:06:12 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Wed Apr 22 13:14:07 2026 -0700" }, "message": "http2: move some server test infrastructure into a common file\n\nIn preparartion for adding a build-tag guarded version of the\nclient and server which wrap net/http\u0027s implementation, move\nvarious bits of server_test.go into a separate file which\nwill be used by both implementations.\n\nFor #78508\n\nChange-Id: I0574e9628847b194177b21d056e4b8d56a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/765380\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "f70faeaf29edebe946e858be8acfebd70276cf33", "tree": "43b90a3a44e4764a153bf86782df71c1d1b67935", "parents": [ "a4ac1da285b790bb85287456145a4dcf3b0d1d4b" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Apr 20 14:15:50 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Mon Apr 20 14:26:00 2026 -0700" }, "message": "Revert \"http2/hpack: prevent HeaderField from escaping during encoding\"\n\nThis reverts commit 9ee1e484e5aab0d95b3babbc6f1384d03f4f9e22.\n\nReason for revert: causes increased heap allocation for workloads with\nhigh dynamic-table churn.\n\nFor golang/go#78833\n\nChange-Id: I30ad85eeb0af3fe2f2b388a9490b955b6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/769100\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "a4ac1da285b790bb85287456145a4dcf3b0d1d4b", "tree": "f1877249487f6002507dc22406b7cce13cfa096b", "parents": [ "0c5a180e97c802e4060ff31d4763bdfadecb653a" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Apr 16 16:01:22 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Apr 16 14:14:19 2026 -0700" }, "message": "http3: adjust HTTP/3 linknamed functions to return quic.Endpoint channels\n\nnet/http tests need access to the QUIC endpoints of HTTP/3 servers and\ntransports in order to synchronously close them at the end of tests.\nTherefore, adjust the signature of the linknamed functions accordingly.\n\nFor golang/go#78737\nFor golang/go#70914\n\nChange-Id: I5df6528b6d1bae8c7fe416f56cd9ce4c6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/768001\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "0c5a180e97c802e4060ff31d4763bdfadecb653a", "tree": "0c9dc0479a93c00f6119470a93f8cbacc6fc927b", "parents": [ "6c6f714fe2e55253602c0286e1e4bddfd571c2b7" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Apr 16 15:48:15 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Apr 16 14:14:07 2026 -0700" }, "message": "internal/http3: add TransportOpts argument to RegisterTransport\n\nRegisterTransport now takes in TransportOpts, which allows\nRegisterTransport user to specify how an HTTP/3 transport should open a\nQUIC endpoint, and what QUIC configuration to use.\n\nPractically, this allows us to access the QUIC endpoint that an HTTP/3\ntransport is using. This allows us to synchronously close QUIC\nconnections while exercising our tests in std, to prevent goroutine\nleaks after running tests.\n\nFor golang/go#78737\nFor golang/go#70914\n\nChange-Id: I34f9756040044624ca9ea21654b249526a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/768000\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "6c6f714fe2e55253602c0286e1e4bddfd571c2b7", "tree": "d63f7b1dd95a1b9cd9021a3ff58d324e4fceb5fd", "parents": [ "cd9d45923a85a063926c715df0a356bb5d69d16a" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Apr 16 09:06:45 2026 -0400" }, "committer": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Apr 16 07:47:27 2026 -0700" }, "message": "http2/h2c: deprecate package\n\nFixes #72039\n\nChange-Id: Ib4a21b79cfdf788988f58c04da2d414c6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/767720\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "cd9d45923a85a063926c715df0a356bb5d69d16a", "tree": "2d24d28933059ed62a02b56b5eeebde262626930", "parents": [ "515a7a7a5ae1159370ac3c0974cca35b1d5a3fd8" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Apr 15 13:58:31 2026 -0400" }, "committer": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Apr 16 07:47:00 2026 -0700" }, "message": "internal/httpcommon: reject :path with no absolute-path\n\nThe :path header should contain an absolute-path production,\nsuch as \"/foo/bar\" or \"//////many-initial-slashes-okay\".\nIt should not contain an absolute URL such as \"http://nope/no\".\n\nFor #78762\n\nChange-Id: Id1697c5c8ff84fdd843361820ba140546a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/767420\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "515a7a7a5ae1159370ac3c0974cca35b1d5a3fd8", "tree": "f28fd86691efba4f2936e8d136fa41fb2ca57feb", "parents": [ "8c4c965e028475082408749b50ed7a686df0d265" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Apr 13 22:02:08 2026 -0400" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 16 06:43:31 2026 -0700" }, "message": "quic: fix source conn ID length check in packet parser\n\nCut-and-paste error: We were checking the DCID length\nrather than the SCID length.\n\nFixes #78703\n\nChange-Id: I0cd7d1d4f2faad560b760ec5bef12bde6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/766944\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Florian Lehner \u003clehner.florian86@gmail.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "8c4c965e028475082408749b50ed7a686df0d265", "tree": "71b4d53ee36f430a35c48a4b456396377e80e281", "parents": [ "de747c49067a073c894c074b33fda61c2d9f36cb" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Tue Apr 14 21:45:19 2026 -0400" }, "committer": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Apr 16 06:28:47 2026 -0700" }, "message": "idna: update from x/text, fix ToUnicode and all-ASCII xn-- labels\n\nChanges produced by running:\n\n\tcd src/golang.org/x/text\n\tgo generate ./...\n\nManually removed a number of files which have been\nremoved from x/text, since the \"go generate\" doesn\u0027t\nremove files.\n\nAmong other changes, this brings in a fix for ToUnicode\nwhich causes it to correctly reject xn-- labels which\ndecode to an all-ASCII label. (The prior behavior\ntechnically followed the UTS 46 algorithm, but that\nalgorithm contained a specification bug corrected in\nUTS 46 revision 33.)\n\nFor golang/go#78760\n\nChange-Id: I79d455d3ed991f2db241954cc93066fc6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/767220\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "de747c49067a073c894c074b33fda61c2d9f36cb", "tree": "fdf0f96190bbce28d2a09299163a0ee093f5cf01", "parents": [ "88ddafc111e48a691110cf4821e70348534b349d" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Apr 13 22:02:08 2026 -0400" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 16 06:24:43 2026 -0700" }, "message": "quic: enforce connection ID limits in Initial packets\n\nReject Initial packets containing a source or destination\nconnection ID which exceeds the protocol limit.\n\nDrop a useless length check in parseGenericLongHeaderPacket. This\nfunction does not enforce the QUICv1 connection ID length limit,\nsince it needs to parse packets from potential future protocol\nversions. The version-independent connection ID length limit\nis 255 bytes, which is the largest length which can by encoded\nby the 1-byte length prefix.\n\n(The source connection ID length limit check was accidentally\nchecking the destination connection ID, but since neither\nconnection ID could ever actually be longer than 255 bytes,\nthis check was never going to succeed anyway.)\n\nRework some test infrastructure to set the testConn\u0027s\npeerConnID automatically when accepting a new server connection.\n\nFor #78703\n\nChange-Id: I527425c2f5111caacb25e9401f8748276a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/766943\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "88ddafc111e48a691110cf4821e70348534b349d", "tree": "9dbda8b2e5a2f2075acfcffcd9c943eaaac9ee37", "parents": [ "eeec9fa0b9d35522d78fd89e86f27ef6af33d79e" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Apr 13 22:02:08 2026 -0400" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 16 06:24:40 2026 -0700" }, "message": "quic: avoid race in benchmarks\n\nFix benchmarks to not exit with a goroutine still running,\nwhich causes racy access to B.N.\n\nChange-Id: If68896f210f2d7e33c5cc7da30e59d676a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/766942\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\n" }, { "commit": "eeec9fa0b9d35522d78fd89e86f27ef6af33d79e", "tree": "9e7b8e6225774bc61808318507f8dffa085d7948", "parents": [ "d7ffc5eda49c464aba55749f0d584ecfdf136571" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Apr 13 22:02:08 2026 -0400" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 16 06:24:37 2026 -0700" }, "message": "quic: fix race condition in fast-path reads\n\nReceiving resent data for a stream could cause racy access to\nthe fast-path input buffer, when the resent data overwrote\nthe input buffer (with identical data).\n\nAdd a mutex guarding the fast-path input buffer.\nThis is not necessary to avoid this race, but makes\nracing calls to Stream.Read safe (so far as the race\ndetector is concerned, at least).\n\nAvoid rewriting portions of the input buffer which\nmight be accessed by the fast-path.\n\nFor #78737\n\nChange-Id: I1cb086e3590e6854935b7c655d1573e16a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/766941\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "d7ffc5eda49c464aba55749f0d584ecfdf136571", "tree": "a1028d402d07320193e08c60a1e29ec237bd6e3c", "parents": [ "cd64923cd7f4f384296109a6ac08cb4936e2e490" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Apr 13 22:02:08 2026 -0400" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 16 06:24:09 2026 -0700" }, "message": "quic: fix race condition in fast-path writes\n\nClosing a stream caused racy access to the fast-path output buffer.\nAdd a mutex guarding the buffer. The mutex is uncontended except\nwhen closing.\n\nFor #78737\n\nChange-Id: I6e964a4564a5fb531305797f8c39eead6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/766940\nLUCI-TryBot-Result: golang-scoped@luci-project-accounts.iam.gserviceaccount.com \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "cd64923cd7f4f384296109a6ac08cb4936e2e490", "tree": "6b4db6bce801be75a0f7bfed3f5e5853a9c93c7d", "parents": [ "a8d1fc14d9e33e1f6842ab78a0127d42cd8fff44" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Apr 10 15:06:25 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Apr 10 12:27:40 2026 -0700" }, "message": "internal/http3: avoid data race for roundTripState.reqBody\n\nThere is currently a data race because roundTripState.reqBody is\nwritten in writeBodyAndTrailer, which might execute concurrently with\ncloseReqBody that reads roundTripState.reqBody.\n\nTo avoid the data race, initialize roundTripState.reqBody early. This\nensures that the roundTripState is no longer going to be written to by\nthe time that it is given as part of a transportResponseBody, whose\nClose method uses closeReqBody.\n\nThe test for this behavior is in std. This data race was found as part\nof running net/http test against x/net/internal/http3.\n\nFor golang/go#70914\n\nChange-Id: I9c97007bea4883e93c2075d1f1b48c356a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/765520\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nAuto-Submit: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "a8d1fc14d9e33e1f6842ab78a0127d42cd8fff44", "tree": "b306b307f7bf7d01dcccdfc4bfd60ae3c4f8aecb", "parents": [ "056ac742146af742aa760d690269c02fa238cc7a" ], "author": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 09 09:03:39 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 09 12:17:33 2026 -0700" }, "message": "go.mod: update golang.org/x dependencies\n\nUpdate golang.org/x dependencies to their latest tagged versions.\n\nChange-Id: I955c34b3a622f9e18ced91a9adcf307dc845607a\nReviewed-on: https://go-review.googlesource.com/c/net/+/764640\nAuto-Submit: Gopher Robot \u003cgobot@golang.org\u003e\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\nReviewed-by: David Chase \u003cdrchase@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "056ac742146af742aa760d690269c02fa238cc7a", "tree": "08efcbddb0c3ece91ee284612065c647f1808b82", "parents": [ "c85f61116e47b1523036c3005f8b2923b661eb64" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Apr 06 13:52:03 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Mon Apr 06 13:47:16 2026 -0700" }, "message": "quic: avoid depending on golang.org/x/sys/unix\n\nTo test our HTTP/3 implementation against tests in std, we need to\nvendor the http3 package. However, http3 relies on quic, which relies on\nx/sys/unix.\n\nVendoring x/sys/unix in std is not allowed (golang/go#32102). Therefore,\nreplace x/sys/unix usage with syscall to avoid the dependency.\n\nFor golang/go#70914\n\nChange-Id: Iad8efff92c9f4135a0d8b89f8544fa206a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/763160\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "c85f61116e47b1523036c3005f8b2923b661eb64", "tree": "041b143107ed41ab3e19b411dd9edb2d89ecd820", "parents": [ "805fc81a196b95c3c00f02e135ffb8a8d5582bdf" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Mar 30 15:37:06 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Apr 02 18:44:39 2026 -0700" }, "message": "http3: add http3 package for testing in std\n\nThis CL adds an initial http3 package that serves as a thin wrapper over\nthe exported functions in the internal/http3 package. Doing so allows us\nto import and vendor the HTTP/3 implementation in x/net from std. This\nwill be used to run std tests against our HTTP/3 implementation.\n\nTo prevent users from unexpectedly using this package, only unexported\nsymbols have been added to the package, with linkname that only allows\nthem to be used by the net/http_test package in std.\n\nFor golang/go#70914\n\nChange-Id: I6b420bb1eb07fd86174ee4c1ede43ef26a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/762041\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "805fc81a196b95c3c00f02e135ffb8a8d5582bdf", "tree": "869d572f7b2159d892413a40dbe9e3060aae543a", "parents": [ "e63b894ab3cd38a1d05396530dccde7ffa3f68d0" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Apr 01 16:00:32 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 02 14:10:21 2026 -0700" }, "message": "http2: add transport API tests\n\nAdd a collection of tests which specifically exercise the user-facing\nclient APIs in this package.\n\nMany of these tests are somewhat redundant with existing tests in\nthe package. Their intent is to ensure consistency between various\nways users may access HTTP/2 features:\n\n - Using a x/net/http2.Transport directly.\n - Using a net/http.Transport configured with ConfigureTransport.\n\nand:\n\n - Using the HTTP/2 implementation in x/net/http2.\n - Using the HTTP/2 implementation in net/http, wrapped by x/net/http2.\n (A case to be implemented in future CLs.)\n\nFor #78508\n\nChange-Id: I228cc1980461ca5c5b4a080bb966450b6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/762320\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "e63b894ab3cd38a1d05396530dccde7ffa3f68d0", "tree": "876383392e86df41b4070d0348a8e7057f7f0b5e", "parents": [ "9ee1e484e5aab0d95b3babbc6f1384d03f4f9e22" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Apr 01 16:00:32 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Apr 02 13:05:29 2026 -0700" }, "message": "http2: support testing via net/http.Transport.RoundTrip\n\nThe testTransport test harness provides support for\ntesting Transport.RoundTrip.\n\nRework testTransport so it can use RoundTrip on either\nx/net/http2.Transport or net/http.Transport.\n\nThis change will let us write tests which verify\nconsistent behavior when using the http2 package\u0027s\nTransport directly or via the net/http Transport.\n\nFor #78508\n\nChange-Id: I7bc34973c0cc9a26b1f0cff56d4fdf4f6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/762300\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "9ee1e484e5aab0d95b3babbc6f1384d03f4f9e22", "tree": "7be45eb6e195e25c757b87647b5bb08e57d1dba2", "parents": [ "1e71bd86e4a302b4e731bc06da6eb51679c7bd49" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Apr 01 16:51:39 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Apr 01 16:06:24 2026 -0700" }, "message": "http2/hpack: prevent HeaderField from escaping during encoding\n\nChange-Id: I8ccd4f01a74d005128c5e2949ca952706a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/762040\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "1e71bd86e4a302b4e731bc06da6eb51679c7bd49", "tree": "9b9a38b202194200be862182fa9fe35067d4640f", "parents": [ "7bca15042b9d2bda1402cb42232a9c6ddbae6212" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Mar 31 15:15:30 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Tue Mar 31 15:18:17 2026 -0700" }, "message": "http2: prevent hanging Transport due to bad SETTINGS frame\n\nThis CL backports https://go.dev/cl/761581 to x/net.\n\nFixes golang/go#78476\nFixes CVE-2026-33814\n\nChange-Id: Ied435a51fdd8664d41dae14d082c39c76a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/761640\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "7bca15042b9d2bda1402cb42232a9c6ddbae6212", "tree": "7516b217afc8833769b21285f4f5b38234410056", "parents": [ "44c41bee5028537e64410b1583e8ae329ceac284" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Mar 30 21:28:20 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Tue Mar 31 15:18:10 2026 -0700" }, "message": "internal/http3: respect net/http Server Shutdown context when shutting down\n\nNow that net/http Server passes down its context upon shutting down,\nmake sure that said context is utilized in our HTTP/3 server shutdown\nimplementation rather than using a fixed timeout.\n\nFor golang/go#70914\nFor golang/go#77440\n\nChange-Id: Ic7e16714ae8ecd8f6e6320fcca107c736a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/761463\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "44c41bee5028537e64410b1583e8ae329ceac284", "tree": "717bc3921d557eb321166ce0b21a094348aa8c43", "parents": [ "228a67a374710bff77fc490e7f538b317c34e247" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Mar 27 16:30:01 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Mar 27 14:17:38 2026 -0700" }, "message": "internal/http3: prevent server from holding mutex when sleeping during shutdown\n\nIf the server holds the lock while sleeping when shutting down,\nconnections will not be able to unregister themselves from s.activeConns\nafter receiving the GOAWAY frame from the server. This will then cause\nthe server to abort connections unnecessarily after it is done sleeping.\n\nChange-Id: I2bce91785db2d138f7bea3a26311139c6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/760560\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "228a67a374710bff77fc490e7f538b317c34e247", "tree": "6ba29012f16c08a35b2a6ece6b66735a386d252b", "parents": [ "7b75446a98c980b353fe5d1c58057d7a7353f1dd" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Mar 09 17:01:30 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Mar 27 14:10:48 2026 -0700" }, "message": "internal/http3: add CloseIdleConnections support in transport\n\nCurrently, our HTTP/3 transport creates a QUIC endpoint that is never\nclosed when dialing a connection for the first time.\n\nTo prevent lingering resource that is never cleaned up until a program\nexits, we now close the QUIC endpoint when CloseIdleConnections is\ncalled and no connections need the endpoint anymore.\n\nFor golang/go#70914\nFor golang/go#77440\n\nChange-Id: I0e4cb131c161216dd8e6abacfc4e57e79be63d7c\nReviewed-on: https://go-review.googlesource.com/c/net/+/754741\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "7b75446a98c980b353fe5d1c58057d7a7353f1dd", "tree": "17dd29b0824a34d8917f504b83548efc0be4010a", "parents": [ "c48da250e237b8355b2943875449ef071bee07bb" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Feb 27 14:50:59 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Mar 26 17:43:53 2026 -0700" }, "message": "internal/http3: add shutdown support for server\n\nCurrently, our HTTP/3 server that has been registered against a net/http\nServer will keep running even after the net/http Server has shutdown.\n\nTherefore, add support for shutting down a running HTTP/3 server, and\nuse it as part of net/http.Server.RegisterOnShutdown.\n\nFor golang/go#70914\nFor golang/go#77440\n\nChange-Id: I5d2acd2cb0fef64ecf5f8d81ae79f3399ed97b00\nReviewed-on: https://go-review.googlesource.com/c/net/+/754580\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "c48da250e237b8355b2943875449ef071bee07bb", "tree": "10db2e292e2b21c9190587d03a2ff78e45d78b69", "parents": [ "a4b5ad55afdef6716207e27b773276973ae98dc9" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Mar 18 15:34:19 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Mar 25 14:16:51 2026 -0700" }, "message": "internal/http3: panic in server handler if given status that is not 3 digits\n\nOur HTTP/1 and HTTP/2 server implementations will panic when WriteHeader\nis called from a handler with a status code that is not 3 digits. For\nconsistency, do this too for HTTP/3.\n\nFor golang/go#70914\n\nChange-Id: I65aca23ea186481ae06c7388db487e476a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/759300\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "a4b5ad55afdef6716207e27b773276973ae98dc9", "tree": "0c45502835edb7e5d9156d439deb4797bdebed2f", "parents": [ "70576f07de22e6581442a2b1b8eedf53d217d7c0" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Mar 24 19:04:14 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Tue Mar 24 17:51:37 2026 -0700" }, "message": "internal/http3: streamline synctest.Wait usage in tests\n\nChange-Id: I40e5e2d033e9f1c51b419de1347295073b99de9a\nReviewed-on: https://go-review.googlesource.com/c/net/+/758824\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "70576f07de22e6581442a2b1b8eedf53d217d7c0", "tree": "8041adb945ed380634639f55fd82e0958c4d4930", "parents": [ "157f4d57d9b24788e919720b16da6c14e951548c" ], "author": { "name": "Olivier Mengué", "email": "olivier.mengue@gmail.com", "time": "Mon Mar 23 09:00:52 2026 +0100" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Tue Mar 24 07:44:58 2026 -0700" }, "message": "http/httpproxy: godoc fixes\n\nDoc improvements:\n- use godoc link to net/http\n- link use go.dev instead of golang.org\n- fix typo in ProxyFunc doc\n\nChange-Id: I7be0ebc78b3c34bc50fac6edc2c20a713478a315\nReviewed-on: https://go-review.googlesource.com/c/net/+/757980\nAuto-Submit: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "157f4d57d9b24788e919720b16da6c14e951548c", "tree": "af31d2b788f4234af0868edd4dcf516dabef503b", "parents": [ "faf99b58f2b2b4fdeb7f313377d58561747cb2a5" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Mar 23 15:23:38 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Mon Mar 23 13:38:59 2026 -0700" }, "message": "quic: prevent server panic when handling Retry packets with short connection ID\n\nPer RFC 9000, a Destination Connection ID in a Retry packet is of valid\nlength as long as it is 20 bytes or shorter. However, when validating\ntokens for Retry packets, our server currently makes an assumption that\nit will only deal with a Destination Connection ID that is exactly 20\nbytes long. As a result, a misbehaving or malicious client can cause our\nserver to panic.\n\nWhen our server validates a token for a Retry packet, the nonce is\npartially constructed from the Destination Connection ID. Due to our\nserver\u0027s bad assumption, a shorter-than-expected Destination Connection\nID will result in cipher.AEAD.Open being given a nonce of bad length. We\ncurrently use XChaCha20-Poly1305 from x/crypto for our AEAD\nimplementation, which will panic when given a nonce of bad length.\n\nFixes golang/go#78292\n\nChange-Id: Ieb65d8b84bcb5c531bdb46d5c75b13dad9c76eb2\nReviewed-on: https://go-review.googlesource.com/c/net/+/758360\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "faf99b58f2b2b4fdeb7f313377d58561747cb2a5", "tree": "2c50cf21fc747645f49dc85571e230cd51755d44", "parents": [ "af27605f5193ce6c0be58ec5662a214755bea7e3" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Mar 23 10:56:53 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Mar 23 11:07:07 2026 -0700" }, "message": "quic: document that this is not yet covered by the security policy\n\nImplied by \"work in progress\", but let\u0027s be explicit about it.\n\nChange-Id: I5cc39129e5af8effb4cc64b0b07ff2176a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/758240\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Roland Shoemaker \u003croland@golang.org\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "af27605f5193ce6c0be58ec5662a214755bea7e3", "tree": "dff3fb67f76380b96492ce81f2c44dca5487957e", "parents": [ "2a1610a02971e7b05256bbfd07c0b44254f884da" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Mar 19 16:50:03 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Mar 19 19:36:52 2026 -0700" }, "message": "quic: avoid usage of net.ReadMsgUDPAddrPort in non-linux and darwin builds\n\nAmongst the non-linux and darwin builds, net.ReadMsgUDPAddrPort is not\nimplemented yet in plan9. There does not seem to be any reason why we\ncould not use net.ReadFromUDPAddrPort instead, which is implemented in\nplan9.\n\nTherefore, use net.ReadFromUDPAddrPort instead for better compatibility.\n\nFixes golang/go#78223\n\nChange-Id: Id397f28368f0cc291b12c6e06df3de64321a623b\nReviewed-on: https://go-review.googlesource.com/c/net/+/757100\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Ethan Lee \u003cethanalee@google.com\u003e\nReviewed-by: Richard Miller \u003cmillerresearch@gmail.com\u003e\n" }, { "commit": "2a1610a02971e7b05256bbfd07c0b44254f884da", "tree": "d53b17e3d82d95ac55d0c25deb2ef4e5ecfaf954", "parents": [ "6d6c88c8948d8c9eb20dc1e3933e660179462ce9" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Mar 19 14:36:31 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Mar 19 11:46:05 2026 -0700" }, "message": "internal/http3: skip TestNetHTTPIntegration on plan9\n\nTestNetHTTPIntegration relies on ReadMsgUDP which is currently\nunimplemented in plan9.\n\nFor golang/go#78223\n\nChange-Id: I958f5f1324236e540a9c8cdf4f2b3c805f023732\nReviewed-on: https://go-review.googlesource.com/c/net/+/757000\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Ethan Lee \u003cethanalee@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "6d6c88c8948d8c9eb20dc1e3933e660179462ce9", "tree": "1720512b8b8c16521ac7160730d5de58e907b0af", "parents": [ "10ac4db83219fc31961ca0a825f15827ffbca5f3" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Mar 19 11:22:58 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Mar 19 10:22:40 2026 -0700" }, "message": "internal/http3: prevent panic in TestNetHTTPIntegration\n\nt.Errorf was accidentally used instead of t.Fatal for an error that we\ncannot tolerate.\n\nFor golang/go#78223\n\nChange-Id: Ice6b566359f8d138627f7a3509b625741b9bd5cf\nReviewed-on: https://go-review.googlesource.com/c/net/+/756920\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Ethan Lee \u003cethanalee@google.com\u003e\nReviewed-by: Richard Miller \u003cmillerresearch@gmail.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "10ac4db83219fc31961ca0a825f15827ffbca5f3", "tree": "cbb3739a64aed9366b5c3205fdf254ace75e3683", "parents": [ "af2121a04e7d6eb35cb7b1069cf3bcedad07d83a" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Fri Mar 13 14:49:14 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Mar 16 10:43:20 2026 -0700" }, "message": "http2: deflake TestServer_Rejects_Too_Many_Streams\n\nThis CL is the x/net counterpart to CL 755320.\n\nThis test contains a race condition in the server handler:\n\n\tinHandler \u003c- streamID\n\t\u003c-leaveHandler\n\nWe assume that all requests queue reading from leaveHandler in order,\nbut it is possible for the second request (stream id 3) to arrive at\nleaveHandler before the first (stream id 1).\n\nWe could fix the race with a judicious synctest.Wait, but rewrite\nthe test to use serverHandlerCall to manipulate server handlers,\nwhich permits us to precisely pick which request to unblock.\n\nFixes #69670\n\nChange-Id: I9507d1dba07f7d62bcdc6c9bb67c47466a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/755081\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "af2121a04e7d6eb35cb7b1069cf3bcedad07d83a", "tree": "0eaa7a1ca71f4389941a56e5363ad8e679087dc6", "parents": [ "316e20ce34d380337f7983808c26948232e16455" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Mar 12 12:33:51 2026 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Mar 12 10:37:40 2026 -0700" }, "message": "internal/http3: avoid blocking when closing a server\u0027s QUIC endpoint\n\nSeveral tests such as TestServerReceivePushStream are currently flaky.\nThis is because the synctest test finishes before the test server\nfinishes closing its QUIC endpoint.\n\nTo prevent this flakiness, make sure that the server uses a canceled\ncontext when closing the QUIC endpoint, so as to not block.\n\nFixes golang/go#78100\nFixes golang/go#78101\nFixes golang/go#78102\n\nChange-Id: I88a257d68dfbcf4e1f18b600a2f11dfdabecb078\nReviewed-on: https://go-review.googlesource.com/c/net/+/754740\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "316e20ce34d380337f7983808c26948232e16455", "tree": "42cc15798386179f3383bdd78b1dd1cc494e1344", "parents": [ "9767a42264fa70b674c643d0c87ee95c309a4553" ], "author": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Wed Mar 11 15:19:38 2026 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Wed Mar 11 17:35:42 2026 -0700" }, "message": "go.mod: update golang.org/x dependencies\n\nUpdate golang.org/x dependencies to their latest tagged versions.\n\nChange-Id: Ie4bfdbe9480a9bcd1384b4f8060239cc7cfc6dee\nReviewed-on: https://go-review.googlesource.com/c/net/+/754560\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Gopher Robot \u003cgobot@golang.org\u003e\nReviewed-by: Cherry Mui \u003ccherryyz@google.com\u003e\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\n" }, { "commit": "9767a42264fa70b674c643d0c87ee95c309a4553", "tree": "bdc3f2952999f791832e05aca34d9bed39795a1e", "parents": [ "4a812844d820f49985ee15998af285c43b0a6b96" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Feb 27 14:49:59 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Mar 11 15:40:04 2026 -0700" }, "message": "internal/http3: add support for plugging into net/http\n\nThis CL adds RegisterServer and RegisterTransport, which are used to\nplug our HTTP/3 implementation into net/http. For now, (inaccurately)\nassume that RegisterTransport do not take any configs for simplicity.\n\nVarious exported symbols have also been changed to be unexported, since\nHTTP/3 will be used only via net/http and will not support direct usage.\n\nFor golang/go#77440\n\nChange-Id: I7b9528c193c18f5049da7f497d4259153b8770eb\nReviewed-on: https://go-review.googlesource.com/c/net/+/739881\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "4a812844d820f49985ee15998af285c43b0a6b96", "tree": "86a205a3175387b0b3bcc9723d14be6825c43d57", "parents": [ "dec6603c16144712aab7f44821471346b35a2230" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Mar 05 16:10:00 2026 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Mar 09 08:35:37 2026 -0700" }, "message": "http2: update docs to disrecommend this package\n\nEvery supported feature of the HTTP/2 client and server\nis accessible via the net/http package. Recommend that\nusers use net/http rather than importing this package.\n\nFor #67810\n\nChange-Id: I58b9c65af27e2f7172af493f96ee589a6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/752160\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "dec6603c16144712aab7f44821471346b35a2230", "tree": "c35e9f437b9f7fb553bc690c91b624c2a42e4a67", "parents": [ "8afa12f927391ba32da2b75b864a3ad04cac6376" ], "author": { "name": "Steven Richards", "email": "steven.richardspc@gmail.com", "time": "Sat Feb 28 15:35:15 2026 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Fri Mar 06 10:52:24 2026 -0800" }, "message": "dns/dnsmessage: reject too large of names early during unpack\n\nThe existing implementation of the Name.unpack method does not check the\nlength of the domain name until parsing is complete. This allows a\nmalicious user to supply an unreasonably large name and wastle cycles\nparsing. This change moves an equivalent check into the loop during\nprocess to short-circuit if we\u0027ve created too large of a name.\n\nFor golang/go#77540\n\nChange-Id: I4c4bf20c0342825a09cefd9b0b3c0bdce0c80137\nReviewed-on: https://go-review.googlesource.com/c/net/+/750260\nReviewed-by: Roland Shoemaker \u003croland@golang.org\u003e\nAuto-Submit: Roland Shoemaker \u003croland@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Cherry Mui \u003ccherryyz@google.com\u003e\n" }, { "commit": "8afa12f927391ba32da2b75b864a3ad04cac6376", "tree": "c286f72ee091850f858fe77b00510770dedd911c", "parents": [ "38019a2dbc2645a4c06a1e983681eefb041171c8" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Feb 18 15:35:29 2026 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Mar 05 16:11:29 2026 -0800" }, "message": "http2: deprecate write schedulers\n\nThe user-provided write scheduler mechanism provides too\nmuch visibility into implementation internals, is difficult\nto use, and limits our ability to improve performance.\n\nFixes golang/go#67817\n\nChange-Id: Iac02c0902f805b671b4541a5dd7eafe76a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/751640\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\n" }, { "commit": "38019a2dbc2645a4c06a1e983681eefb041171c8", "tree": "7107669c74e45e7f4557456da95a6b7f0ef20a9c", "parents": [ "039b87fac41ca283465e12a3bcc170ccd6c92f84" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Mar 04 14:09:29 2026 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Wed Mar 04 14:19:29 2026 -0800" }, "message": "http2: add missing copyright header to export_test.go\n\nChange-Id: I7add2f2ef54ef6f985154a60399768e76a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/751620\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "039b87fac41ca283465e12a3bcc170ccd6c92f84", "tree": "0846f463c10e5aa1375133ce900827c3b035429e", "parents": [ "6267c6c4c825a78e4c9cbdc19c705bc81716597c" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Mar 03 13:07:09 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Mar 04 06:38:46 2026 -0800" }, "message": "internal/http3: return error when Write is used after status 304 is set\n\nIn our HTTP/1 and HTTP/2 implementations, calling Write in a server\nhandler after WriteHeader has been called with status 304 will return an\nhttp.ErrBodyNotAllowed error. This change adds the same behavior for the\nHTTP/3 server.\n\nFor golang/go#70914\n\nChange-Id: I6be926412d51217a8b88b2ad4ce79935dd3e7af7\nReviewed-on: https://go-review.googlesource.com/c/net/+/751140\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "6267c6c4c825a78e4c9cbdc19c705bc81716597c", "tree": "70f7805740d8c864096c9c8f6d5dc26ca34e288f", "parents": [ "591bdf35bce56ad50f53555c3cbb31e4bdda2d58" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Feb 25 18:28:18 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Mar 04 06:38:43 2026 -0800" }, "message": "internal/http3: add HTTP 103 Early Hints support to ClientConn\n\nRoundTrip will now call httptrace.ClientTrace.Got1xxResponse, if any,\nwhen receiving 1xx status response from a peer. This allows our client\nand server to use HTTP 103 end-to-end.\n\nGot100Continue and Wait100Continue have also been added to RoundTrip as\nthey are nearby. The rest of httptrace.ClientTrace will be added in the\nfuture.\n\nFor golang/go#70914\n\nChange-Id: Ia7ef7dd026a5390225149da3d76b06a2a372c009\nReviewed-on: https://go-review.googlesource.com/c/net/+/749265\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "591bdf35bce56ad50f53555c3cbb31e4bdda2d58", "tree": "fd4bc6c3ca2ec6e9195719812c3892b31de97fd8", "parents": [ "1faa6d8722697d9a1d8d4e973b3c46c7a5563f6c" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Feb 25 17:36:02 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Mar 04 06:38:40 2026 -0800" }, "message": "internal/http3: add HTTP 103 Early Hints support to Server\n\nUsing WriteHeader from Server handler to send status 103 will now work\nsimilarly to HTTP/1 and HTTP/2. Additionally, the code path for handling\ninformational response header in general has also been simplified.\n\nSupport for the client is added in a separate change.\n\nFor golang/go#70914\n\nChange-Id: I07bd15fc56d0b1e18f1831c9002d548dbfb5beb5\nReviewed-on: https://go-review.googlesource.com/c/net/+/749264\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "1faa6d8722697d9a1d8d4e973b3c46c7a5563f6c", "tree": "857ef15457938e1fb32bab91a54a7b0c455033d9", "parents": [ "8d297f1cac94a449ef60461f8242b1982bdeb0bc" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Mar 03 17:29:53 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Mar 04 06:38:26 2026 -0800" }, "message": "internal/http3: avoid potential race when aborting RoundTrip\n\nTestRoundTripRequestBodyErrors/read_error currently has a small chance\nof failing due to a race condition.\n\nMost of the time, within writeBodyAndTrailer which runs concurrently,\nwhen a request body returns an error on Read, rt.abort will be called\nwith the Read error. Since rt.abort is idempotent, the Read error will\nthen be propagated and the test would pass as expected.\n\nHowever, before rt.abort is called with the Read error,\nrt.reqBodyWriter.Close is first called, which will close the write\ndirection of the QUIC stream. This creates a small chance where the main\ngoroutine which runs RoundTrip can encounter an EOF error due to the\nclosed stream, and subsequently call rt.abort with a different error\nthan expected, causing test flakiness.\n\nTo fix this, this change makes sure that rt.abort is immediately called\nwhen a request body Read returns an error. This should be a no-op, since\na body Read error has always been prioritized over a potential error\nfrom rt.reqBodyWriter.Close anyways.\n\nFixes golang/go#77782\n\nChange-Id: I9ec5acf82dd95ac91963ce5c8ddbf08b49af9508\nReviewed-on: https://go-review.googlesource.com/c/net/+/751280\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "8d297f1cac94a449ef60461f8242b1982bdeb0bc", "tree": "00d90e6e9ea1e50e58b11aea6ef469ec0a4b0a29", "parents": [ "3eb9327ec1a13119426967d4839e7481f5964615" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Wed Feb 25 09:41:57 2026 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Tue Mar 03 17:16:54 2026 -0800" }, "message": "http2: Move most tests from the http2 package to the http2_test package.\n\nThis change makes it easier to move x/net/http2 into std.\nMoving the http2 package into std and importing it from net/http\n(rather than bundling it as net/http/h2_bundle.go) requires\nremoving the http2-\u003enet/http dependency. Moving tests into\nthe http2_test package allows them to continue importing net/http\nwithout creating a cycle.\n\nChange-Id: If0799a94a6d2c90f02d7f391e352e14e6a6a6964\nReviewed-on: https://go-review.googlesource.com/c/net/+/749280\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\n" }, { "commit": "3eb9327ec1a13119426967d4839e7481f5964615", "tree": "ea93e7eb5af686c37e7fe084756075f9d3a6d3f1", "parents": [ "60b3f6f8ce12def82ae597aebe9031753198f74d" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Mar 02 17:31:25 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Mon Mar 02 15:02:12 2026 -0800" }, "message": "http2: do not retry RoundTrip after peer sends a stream protocol error\n\nYears back, our HTTP/2 server implementation had a stream accounting bug\nthat would cause it to improperly report a PROTOCOL_ERROR. In response\nto this, we modified our Transport to retry RoundTrip when a RST_STREAM\nwith PROTOCOL_ERROR was received from a peer.\n\nAt this point, this retry logic had outlived its usefulness. Instead, it\nmight cause issues, e.g. a client that sends a malformed request will\nkeep retrying repeatedly, despite there being zero chance for the\nrequest to actually succeed.\n\nFixes golang/go#77843\n\nChange-Id: Ic043723e3535f68f91db33d8f6bcd7fc2dbce856\nReviewed-on: https://go-review.googlesource.com/c/net/+/750720\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "60b3f6f8ce12def82ae597aebe9031753198f74d", "tree": "8907839a961f9a0a306d11c0fb21b7f8298e0a20", "parents": [ "b0ca4561757b944abd31a55aa4dccec65dae1847" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Feb 24 18:40:55 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Tue Feb 24 16:19:02 2026 -0800" }, "message": "internal/http3: prevent Server handler from writing longer body than declared\n\nLike in HTTP/1 and HTTP/2, this CL changes the Server handler so that\nWrite will return an error and automatically trim the content being\nwritten, when attempting to write a body content that is longer than\nwhat has been declared in the \"Content-Length\" header.\n\nFor golang/go#70914\n\nChange-Id: I638efb8ec96f926f86c389f3cac79f1b38a5b36b\nReviewed-on: https://go-review.googlesource.com/c/net/+/748681\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "b0ca4561757b944abd31a55aa4dccec65dae1847", "tree": "ea3d3c1a0b5156e95a2101e3423c8a67c584b7e7", "parents": [ "1558ba78062172d9d1f7854c522b74ae29b35c20" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Feb 24 18:28:06 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Tue Feb 24 16:18:58 2026 -0800" }, "message": "internal/http3: fix Write in Server Handler returning the wrong value\n\nWhen changing the approach used in go.dev/cl/746760 (i.e. adjusting\nslices, rather than keeping track of active part of the slices in\nWrite), the return value was unfortunately not updated properly.\n\nThis CL fixes the problem and modifies the test to also verify that\nWrite returns the correct value.\n\nFor golang/go#70914\n\nChange-Id: Ic7dda43d47a85c74c03cf02208a0cd461b58dcf2\nReviewed-on: https://go-review.googlesource.com/c/net/+/748680\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "1558ba78062172d9d1f7854c522b74ae29b35c20", "tree": "a57addda201f5147fdfad7210bcce46f0253d703", "parents": [ "4e1c745a707af4b9a56e5ae2a6805a99df5da1a6" ], "author": { "name": "Sean Liao", "email": "sean@liao.dev", "time": "Sat Feb 07 13:09:20 2026 +0000" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Tue Feb 24 09:08:04 2026 -0800" }, "message": "publicsuffix: update to 2026-02-06\n\nTest cases for wildcards (*) dropped as the\nfollowing registrations changed.\n\n.uberspace.de removal\nhttps://github.com/publicsuffix/list/pull/2540\n\n.bd registration update to replace *.bd\nhttps://github.com/publicsuffix/list/pull/2623\n\nFixes golang/go#77445\n\nChange-Id: Ib3e120e041bcefce4d89b7f8fd4eb5ab752610c5\nReviewed-on: https://go-review.googlesource.com/c/net/+/742285\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: David Chase \u003cdrchase@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "4e1c745a707af4b9a56e5ae2a6805a99df5da1a6", "tree": "ca4de75bfc17597d63990e2af3a4702ebd2c8339", "parents": [ "19f580fd686a6bb31d4af15febe789827169bc26" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Feb 11 18:12:24 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Mon Feb 23 16:04:38 2026 -0800" }, "message": "internal/http3: make Server response include headers that can be inferred\n\nThis CL makes Server automatically set certain headers in a response\nwhen they can be inferred and have not been explicitly set in a handler.\n\nFor now, \"Date\" and \"Content-Type\" headers are supported. Note that\nunlike HTTP/1 and HTTP/2, \"Content-Length\" header is not automatically\nset.\n\nFor golang/go#70914\n\nChange-Id: Ie8d64ee0efb10118dfcba7a156f6f6886f565046\nReviewed-on: https://go-review.googlesource.com/c/net/+/746760\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "19f580fd686a6bb31d4af15febe789827169bc26", "tree": "25f0353941924a735df4f83962097a082562b284", "parents": [ "818aad7ad4e47b7f3a6b94e4145edb6001460ea2" ], "author": { "name": "Konnyaku", "email": "beifengxuanxiao@126.com", "time": "Tue Feb 17 21:23:54 2026 +0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Tue Feb 17 11:57:01 2026 -0800" }, "message": "http2: fix nil panic in typeFrameParser for unassigned frame types\n\nThe addition of FramePriorityUpdate (0x10) in RFC 9218 introduced a gap\nin the frameParsers array indices (0x0a-0x0f). These indices were\ninitialized to nil, causing a panic when typeFrameParser accessed them\nfor unassigned frame types (e.g., ALTSVC 0x0a).\n\nThis change adds a nil check in typeFrameParser to safely fallback to\nparseUnknownFrame for these unassigned types, preventing the crash.\n\nFixes golang/go#77652\n\nChange-Id: I14d7ad85afc1eafabc46417a9fff10f9e0a22446\nReviewed-on: https://go-review.googlesource.com/c/net/+/746180\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Mark Freeman \u003cmarkfreeman@google.com\u003e\n" }, { "commit": "818aad7ad4e47b7f3a6b94e4145edb6001460ea2", "tree": "0c80765dc8339a88736de92f84056280160e7493", "parents": [ "c1bbe1a459794139a79887003b1231d55cf90af7" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Feb 10 15:19:09 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Feb 12 14:10:23 2026 -0800" }, "message": "internal/http3: add server to client trailer header support\n\nThis change implements support for Server to send trailer headers, and\nfor ClientConn to receive said trailer headers. This is just like\ngo.dev/cl/743600, but in the opposite direction.\n\nThe bulk of the implementation relies on the trailer header encoding and\ndecoding support that was added to bodyWriter and bodyReader\nrespectively in go.dev/cl/743600.\n\nFor golang/go#70914\n\nChange-Id: I0efded4b1ac3e3c6b9479f18402e02e9e764d4a2\nReviewed-on: https://go-review.googlesource.com/c/net/+/744220\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "c1bbe1a459794139a79887003b1231d55cf90af7", "tree": "0acb91c932823f48ef48ed26727c34d6b93619de", "parents": [ "29181b8c03a8e33d784696b8cf368d3d7b576a9e" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Feb 09 16:35:46 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Feb 12 14:10:11 2026 -0800" }, "message": "internal/http3: add client to server trailer header support\n\nThis change adds support for ClientConn to send trailer headers, and for\nServer to receive trailer headers. Support for the other direction will\nbe done in a follow-up change.\n\nThe encoding and decoding of trailer headers are done using bodyWriter\nand bodyReader respectively, as these logic are agnostic to whether the\ntrailer header is sent by the client or the server.\n\nFor golang/go#70914\n\nChange-Id: I646d193ae1bc44ddea69b8397d4473d3b11eddf2\nReviewed-on: https://go-review.googlesource.com/c/net/+/743600\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "29181b8c03a8e33d784696b8cf368d3d7b576a9e", "tree": "3675bed9bf509fab7f53e66d9da1ed7410f71477", "parents": [ "81093053d19331b32808127ca215008e61e79b56" ], "author": { "name": "Dmitri Shuralyov", "email": "dmitshur@golang.org", "time": "Wed Feb 11 23:30:32 2026 -0500" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Feb 12 07:53:52 2026 -0800" }, "message": "all: remove go1.25 and older build constraints\n\nNow that the x/net module requires Go 1.25.0,\nthe go1.25 build constraint is always satisfied.\nSimplify the code accordingly.\n\nChange-Id: I3d6fe4a132a26918455489b998730b494f5273c4\nReviewed-on: https://go-review.googlesource.com/c/net/+/744800\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Dmitri Shuralyov \u003cdmitshur@golang.org\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\n" }, { "commit": "81093053d19331b32808127ca215008e61e79b56", "tree": "41a4afc0c2666275909eec5617f3f87e0f30f0bd", "parents": [ "0b37bdfdf0ade471acecbe8410069a34bf3d8fce" ], "author": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Wed Feb 11 18:54:06 2026 +0000" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Wed Feb 11 11:12:51 2026 -0800" }, "message": "all: upgrade go directive to at least 1.25.0 [generated]\n\nBy now Go 1.26.0 has been released, and Go 1.24 is no longer supported\nper the Go Release Policy (see https://go.dev/doc/devel/release#policy).\n\nSee go.dev/doc/godebug#go-125 for GODEBUG changes relevant to Go 1.25.\n\nFor golang/go#69095.\n\n[git-generate]\n(cd . \u0026\u0026 go get go@1.25.0 \u0026\u0026 go mod tidy)\n\nChange-Id: I15bcff18ee591756454a56b212d91aa0453e8e7c\nReviewed-on: https://go-review.googlesource.com/c/net/+/744580\nReviewed-by: Junyang Shao \u003cshaojunyang@google.com\u003e\nAuto-Submit: Gopher Robot \u003cgobot@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Michael Pratt \u003cmpratt@google.com\u003e\n" }, { "commit": "0b37bdfdf0ade471acecbe8410069a34bf3d8fce", "tree": "e9896c668badfae967432ac69dfd1d0ea6d517d0", "parents": [ "9095c1cb86eb20ab69bc26b5484d7ccb4e8c1620" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Feb 09 11:27:09 2026 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Feb 09 11:37:01 2026 -0800" }, "message": "quic: don\u0027t run TestStreamsCreateConcurrency in synctest bubble\n\nThis test creates a pair of real UDP sockets; synctest doesn\u0027t work\nwell with real network connections and interactions between the\nreal network and synctest are probably causing flakiness.\n\nFixes golang/go#76510\n\nChange-Id: I9e8a959e7d0077d8191ceb8284f6c338dd2688ab\nReviewed-on: https://go-review.googlesource.com/c/net/+/743402\nAuto-Submit: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "9095c1cb86eb20ab69bc26b5484d7ccb4e8c1620", "tree": "292d6009b1ceff25c87750de86e7bae73d228d66", "parents": [ "ebddb99633e0fc35d135f62e9400678492c1d3be" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Feb 06 19:09:43 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Mon Feb 09 11:17:18 2026 -0800" }, "message": "internal/http3: more robust handling of request \u0026 response with no body\n\nIn HTTP/3, zero or more DATA frames can come after a HEADERS frame to\nrepresent a request or response body. Our current implementation can\nbehave rather badly when zero DATA frame is sent.\n\nClientConn does not close the write direction of the stream when it has\nno body to send. As a result, our Server can end up reading the next\nframe after a HEADERS frame, only to hang infinitely until the timeout\nis reached. To fix this, when there is no body to send, ClientConn now\ncloses the write direction of the stream as soon as it has finished\nwriting its HEADERS frame. Server will also prevent itself from reading\nthe stream if a Content-Length header with the value 0 is received.\n\nIn the opposite direction (client reading response from a server), a\nsimilar problem also exists, with a slight variant. While our Server\nreliably closes its write direction of the stream as soon as the server\nhandler exits, a problem can still occur when a client receives an empty\nresponse body due to sending a HEAD request. In this case, if the client\ndecides to read the response body, bodyReader might throw an error due\nto a mismatch between the Content-Length header given by the server and\nthe actual body length. This is fixed by making ClientConn aware that\nHEAD requests will always result in an empty response body.\n\nFor golang/go#70914\n\nChange-Id: I1e8970672e7076c9dbf84aec8808632d04bac807\nReviewed-on: https://go-review.googlesource.com/c/net/+/742960\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "ebddb99633e0fc35d135f62e9400678492c1d3be", "tree": "e55a5cca61d140a42d836d4480a09f2b45591993", "parents": [ "4a490d4f5331699af2b0987af2e4cea9925f65af" ], "author": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Feb 09 09:03:18 2026 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Feb 09 09:11:27 2026 -0800" }, "message": "go.mod: update golang.org/x dependencies\n\nUpdate golang.org/x dependencies to their latest tagged versions.\n\nChange-Id: I5e9ca4c38dd59f5c3332ee2e8bac98c338620b69\nReviewed-on: https://go-review.googlesource.com/c/net/+/743400\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\nReviewed-by: Cherry Mui \u003ccherryyz@google.com\u003e\nAuto-Submit: Gopher Robot \u003cgobot@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "4a490d4f5331699af2b0987af2e4cea9925f65af", "tree": "8d86e1b3ea182de3dd58a602e2e46302fcd4dae3", "parents": [ "73fe7011adb0a690693f0318325c60d477c7bb3a" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Feb 05 16:45:11 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Feb 06 16:20:02 2026 -0800" }, "message": "internal/http3: add Expect: 100-continue support to ClientConn\n\nWhen sending a request containing the \"Expect: 100-continue\" header,\nClientConn.RoundTrip will now only send the request body after receiving\nan HTTP 100 status response from the server.\n\nFor golang/go#70914\n\nChange-Id: Ib3acea68b078486bda96426952897c3f2d51b47b\nReviewed-on: https://go-review.googlesource.com/c/net/+/742540\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "73fe7011adb0a690693f0318325c60d477c7bb3a", "tree": "ed27ea00a4a71cab7ac645d8ceb49241b503d031", "parents": [ "af0c9df79dbe35108ead105e4ca28133551e4d1a" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Feb 05 16:12:40 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Feb 06 16:19:57 2026 -0800" }, "message": "internal/http3: add Expect: 100-continue support to Server\n\nWhen serving a request containing the \"Expect: 100-continue\" header,\nServer will now send an HTTP 100 status automatically if the request\nbody is read from within the server handler.\n\nFor golang/go#70914\n\nChange-Id: Ib8185170deabf777a02487a1ded6671db720df51\nReviewed-on: https://go-review.googlesource.com/c/net/+/742520\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "af0c9df79dbe35108ead105e4ca28133551e4d1a", "tree": "d1051b0a0d5ce0c1870b9f923cc042aa84f09808", "parents": [ "e02fb33933734f5c17f8474ef9f54461e776d119" ], "author": { "name": "Carlana", "email": "mail@carlana.net", "time": "Thu Jan 22 14:17:38 2026 +0000" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Fri Feb 06 09:29:02 2026 -0800" }, "message": "html: add NodeType.String() method\n\nUses stringer to add String method to html.NodeType.\n\nFixes golang/go#77006\n\nChange-Id: I8120215f859f4c3e6f17757e106cff3936bfa1ba\nGitHub-Last-Rev: 2082ce04b34a034bbd73f0ac103ab470bfb8d039\nGitHub-Pull-Request: golang/net#244\nReviewed-on: https://go-review.googlesource.com/c/net/+/738100\nReviewed-by: Alan Donovan \u003cadonovan@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Michael Knyszek \u003cmknyszek@google.com\u003e\nAuto-Submit: Alan Donovan \u003cadonovan@google.com\u003e\n" }, { "commit": "e02fb33933734f5c17f8474ef9f54461e776d119", "tree": "ec2f7f2e8426aa81fb786c0e482389cf08ddbd53", "parents": [ "da558ff100e05eb3fd3c94d2f978c062edc070a2" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Feb 04 19:37:39 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Feb 05 18:42:21 2026 -0800" }, "message": "internal/http3: make responseWriter.Flush write headers if not done yet\n\nIn net/http and x/net/http2, flushing an http.ResponseWriter will also\ntrigger headers to be written if none has been written yet at that\npoint. This behavior is sometimes relied on when implementing streaming\nresponses (see http://go.dev/cl/4552062); therefore, let\u0027s do this in\nx/net/internal/http3 too.\n\nFor golang/go#70914\n\nChange-Id: Ib914afc85df21a1cddd1d5019311d3f25d943f8a\nReviewed-on: https://go-review.googlesource.com/c/net/+/742160\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "da558ff100e05eb3fd3c94d2f978c062edc070a2", "tree": "4dc849e20aa437be7d5d4b717bc7a3ac70a6ff74", "parents": [ "d7c76faf077586c8657a8bdd404484c090764e2b" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Feb 04 14:27:11 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Feb 05 18:42:11 2026 -0800" }, "message": "internal/http3: ensure bodyReader cannot be read after being closed\n\nClosing bodyReader currently only closes the underlying QUIC stream. As\na result, any unread data that was written to bodyStream prior to it\nbeing closed can still be read. This is inconsistent with how we expect\nnet/http.Response.Body to behave.\n\nFor golang/go#70914\n\nChange-Id: I58226c0d23ea3bbd97f3ceb5c3659e91660f84c5\nReviewed-on: https://go-review.googlesource.com/c/net/+/741982\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "d7c76faf077586c8657a8bdd404484c090764e2b", "tree": "4504295778d93dcc67e2e462555cbafb8722a832", "parents": [ "64b3af9625ad94ce3f68ac1ab80733f691a49319" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Feb 03 13:03:07 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Tue Feb 03 11:53:50 2026 -0800" }, "message": "internal/http3: make responseWriter behave closer to other http.ResponseWriter\n\nWhile running net/http tests against our HTTP/3 implementation locally,\nsome tests fail due to slight behavior differences in responseWriter\ncompared to other http.ResponseWriter implementations:\n\n- responseWriter does not return a 200 OK response if a server handler\n is completely empty.\n- responseWriter does not have a Flush method, and therefore does not\n implement http.Flusher.\n\nThere are surely more differences, but these are straightforward to fix\nright now.\n\nFor golang/go#70914\n\nChange-Id: Ieb729a4de4ccb55d670eac2369e73c240b9ac8f8\nReviewed-on: https://go-review.googlesource.com/c/net/+/741720\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "64b3af9625ad94ce3f68ac1ab80733f691a49319", "tree": "ce94e3eb23dfed370cc91fc732a5c0d091f8d365", "parents": [ "1973e8da2d578cd964f4c1df41ed8c553f1e17b7" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Jan 28 18:27:34 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Mon Feb 02 09:54:04 2026 -0800" }, "message": "http2: prevent transport deadlock due to WINDOW_UPDATE that exceeds limit\n\nTransport currently deadlocks when receiving a WINDOW_UPDATE for a\nnon-zero stream that increments its window beyond the 2^31-1 bytes\nlimit.\n\nThis is because endStreamError is called to end the non-zero stream,\nwhich tries to lock an already-locked mutex. Therefore, create and use\nendStreamErrorLocked instead, which assumes the mutex is already locked.\n\nFixes golang/go#77331\n\nChange-Id: Iea212f49a1f305d1bddefb8831dbaca00840870c\nReviewed-on: https://go-review.googlesource.com/c/net/+/739700\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "1973e8da2d578cd964f4c1df41ed8c553f1e17b7", "tree": "f6128ec46727ccafcf535ae100668a569facca55", "parents": [ "57ea86db083ff804aa7dfca3489c0f965ab3b0da" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Jan 28 21:47:11 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Jan 29 09:40:02 2026 -0800" }, "message": "internal/http3: add Server support for handling HEAD requests\n\nWhen handling HEAD requests, Server will now always send an HTTP\nresponse with an empty body, even if the HTTP handler would normally\nwrite to the body.\n\nFor golang/go#70914\n\nChange-Id: Id656a8f9901b97357a343e204761f6a47f87f9fe\nReviewed-on: https://go-review.googlesource.com/c/net/+/740160\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "57ea86db083ff804aa7dfca3489c0f965ab3b0da", "tree": "b59055bd2dcc6a733d7e4bf4067c19615020dab5", "parents": [ "bcf2273e0161cfdff7e07ae525773a616fd74daf" ], "author": { "name": "Tobias Klauser", "email": "tklauser@distanz.ch", "time": "Wed Jan 28 11:15:16 2026 +0100" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Jan 29 07:04:13 2026 -0800" }, "message": "icmp, internal/socket, ipv4, ipv6: use binary.NativeEndian\n\nInstead of duplicating it as internal/socket.NativeEndian.\n\nChange-Id: I7c7d76cc561b0f3083e8f33db7b90e2f421c12d5\nReviewed-on: https://go-review.googlesource.com/c/net/+/739840\nAuto-Submit: Tobias Klauser \u003ctobias.klauser@gmail.com\u003e\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Ian Lance Taylor \u003ciant@golang.org\u003e\n" }, { "commit": "bcf2273e0161cfdff7e07ae525773a616fd74daf", "tree": "0334938b50bed97421cbee75aef2cd75e3d9b61d", "parents": [ "6b849cc1221162a8dd99741bdfb4bee86552905a" ], "author": { "name": "Neal Patel", "email": "nealpatel@google.com", "time": "Tue Jan 27 21:33:17 2026 +0000" }, "committer": { "name": "Neal Patel", "email": "nealpatel@google.com", "time": "Tue Jan 27 14:57:05 2026 -0800" }, "message": "quic: clean up impossible nil conditions\n\nChange-Id: I755d937332f671a7f5bedc73aaff31e6fbf07159\nReviewed-on: https://go-review.googlesource.com/c/net/+/739760\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "6b849cc1221162a8dd99741bdfb4bee86552905a", "tree": "277a950d76f93c103f154a90afaec18b21692a9b", "parents": [ "f2078620ee331682e028d9b217f14fb09020811d" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Jan 23 15:14:30 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Jan 23 12:44:29 2026 -0800" }, "message": "internal/http3: populate some http.Request fields\n\nCurrently, we are passing a very barebone http.Request to the server\nhandler: we only initialize an empty http.Request and put whatever info\nwe can get while decoding QPACK headers.\n\nUnfortunately, this causes the Server to panic when parsing requests\nwhose headers are meant to be written to http.Request.URL, as\nhttp.Request.URL was never initialized.\n\nTherefore, make sure that http.Request.URL is initialized. Also,\npopulate other http.Request fields that we can easily figure out as of\nnow.\n\nFor golang/go#70914\n\nChange-Id: Ie6552d6678b430fe4b51069616c0e366791c4e34\nReviewed-on: https://go-review.googlesource.com/c/net/+/738880\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "f2078620ee331682e028d9b217f14fb09020811d", "tree": "a46387e36838f670b1b354207de05009486ff91c", "parents": [ "664045486926d2c0fc255005b0a06e8a4066da85" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Jan 21 15:35:08 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Thu Jan 22 14:59:15 2026 -0800" }, "message": "http2: allow prioritization to be disabled using DisableClientPriority field\n\nAs part of adding support for HTTP/2 stream prioritization, a\nDisableClientPriority field will be added to http.Server to allow users\nto completely disable client prioritization if desired. When\nDisableClientPriority is set to true, HTTP/2 server will revert back to\nthe old behavior where streams are processed in a round-robin manner.\n\nFor golang/go#75500\n\nChange-Id: Ida083b3ac17a953e5ddb3ad7ab8a81f9cde2bfc1\nReviewed-on: https://go-review.googlesource.com/c/net/+/737521\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "664045486926d2c0fc255005b0a06e8a4066da85", "tree": "dc81290370f319b11f1de418ac9c57c7dffe856c", "parents": [ "c082574f2d0c659486e765abc4b518e2e397607c" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Wed Jan 21 17:49:14 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Jan 21 14:58:16 2026 -0800" }, "message": "internal/http3: break dependency on internal/quic/quicwire\n\ninternal/http3 currently depends on internal/quic/quicwire only for its\nSizeVarint when writing SETTINGS frame. This makes bundling\ninternal/http3 into std more complicated since it forces us to also\nbundle internal/quic/quicwire and do import remapping. This CL breaks\nthis dependency for easier bundling.\n\nDependency on internal/quic/quicwire in test codes have been left alone\nsince they do not affect bundling into std, and uses more than just\nSizeVarint.\n\nChange-Id: I2b85c5487ae6cce422deb4509f68f932d3f0de6b\nReviewed-on: https://go-review.googlesource.com/c/net/+/738060\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "c082574f2d0c659486e765abc4b518e2e397607c", "tree": "546cc14a4051f1abba34286f84f4537b83b1ff89", "parents": [ "8f003b3712faff75138744ff08334dddbac0c573" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Jan 19 00:24:51 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Tue Jan 20 09:02:49 2026 -0800" }, "message": "internal/http3: remove GODEBUG\u003dasynctimerchan\u003d0 that is no longer needed\n\nChange-Id: I86e83567b1d07fbc96ac98e0a85eb80748512eef\nReviewed-on: https://go-review.googlesource.com/c/net/+/737520\nReviewed-by: Carlos Amedee \u003ccarlos@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "8f003b3712faff75138744ff08334dddbac0c573", "tree": "1445b86dd06850c5ecf577a789cd00f3569104fc", "parents": [ "8a4d9c198fda8705a57733d419c578acf877aaa4" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Nov 25 13:52:06 2025 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Jan 14 12:33:57 2026 -0800" }, "message": "http2: support SETTINGS_NO_RFC7540_PRIORITIES in SETTINGS frame\n\nTo make sure that clients do not unnecessarily send RFC 7540 priority\nsignals when it would be treated as a no-op, this change makes it so\nthat our server always sends SETTINGS_NO_RFC7540_PRIORITIES in our\nSETTINGS frame when our write scheduler is set to anything other than\nthe RFC 7540 write scheduler.\n\nFor golang/go#75500\n\nChange-Id: I7a54251022087319999deda7efb663f8b251aa95\nReviewed-on: https://go-review.googlesource.com/c/net/+/729141\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "8a4d9c198fda8705a57733d419c578acf877aaa4", "tree": "c4331492c64cfe7b3fa0ca2521b9be45a4812e3e", "parents": [ "a475fa814131e3f2fab5293c68a1619d994e1133" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Nov 25 17:15:58 2025 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Jan 14 12:33:43 2026 -0800" }, "message": "http2: only make streams non-incremental if clients know of RFC 9218\n\nIn RFC 9218, streams are non-incremental by default, meaning that they\nare processed one-by-one to completion. This behavior is the opposite of\nour current default of handling streams in a round-robin manner.\n\nThis might cause a surprising behavior change once we make the RFC 9218\npriority scheduler the default write scheduler for most users (we assume\nthat most users will not be sending RFC 9218 priority signals, at least\ninitially). To avoid surprising users with such a behavior change, this\nCL makes it so that the streams are only made non-incremental once there\nhas been a clear signal that the end-user is aware of RFC 9218.\n\nFor golang/go#75500\n\nChange-Id: Ibd22cb279c43de0190962904c3809007447a5fe3\nReviewed-on: https://go-review.googlesource.com/c/net/+/729140\nReviewed-by: Michael Knyszek \u003cmknyszek@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "a475fa814131e3f2fab5293c68a1619d994e1133", "tree": "10a66c4a41a582644909e9988719b317b18abcbe", "parents": [ "f40205b5b5e758dca22ed35eea3be08720d1a058" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Nov 25 13:52:24 2025 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Jan 14 12:33:32 2026 -0800" }, "message": "http2: add support for setting RFC 9218 priority via header field\n\nRFC 9218 allows HTTP/2 stream priority to be set in two ways: via\nPRIORITY_UPDATE frame and via header field. This change adds support for\nthe latter method.\n\nAs part of supporting priority adjustment via header field, this CL\nalso makes sure to look for the existence of an intermediary. If an\nintermediary exists, default priority will be used for all streams to\nensure fairness between multiple clients who could be using the same\nintermediary.\n\nFor golang/go#75500\nFor golang/go#75936\n\nChange-Id: I6dc409b650fd52fa192d771a16b7a4ac5e51c9aa\nReviewed-on: https://go-review.googlesource.com/c/net/+/729120\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "f40205b5b5e758dca22ed35eea3be08720d1a058", "tree": "bdbbba33ce83326dd0dd84630f99fcb25bec8602", "parents": [ "15b99c1fb8b76f00f3b0ad4572019943173529fc" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Oct 06 16:20:23 2025 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Wed Jan 14 12:33:16 2026 -0800" }, "message": "http2: add initial support for PRIORITY_UPDATE frame defined in RFC 9218\n\nThis change adds initial support for the PRIORITY_UPDATE frame\nintroduced in RFC 9218.\n\nClients can now use a new exported function to write PRIORITY_UPDATE\nframes easily. However, sending PRIORITY_UPDATE frames to the server\ndoes not currently cause any behavior changes: we only use\nPRIORITY_UPDATE frames to adjust stream priority when the RFC 9218 write\nscheduler is being used for a particular connection. However, this\nscheduler is not currently usable yet from any configuration surfaces\nexposed to the user.\n\nFor golang/go#75500\n\nChange-Id: Ie2c821cb0d2faa6e942e209e11638f190fc98e2b\nReviewed-on: https://go-review.googlesource.com/c/net/+/705917\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "15b99c1fb8b76f00f3b0ad4572019943173529fc", "tree": "889986ea96d5876de246446c0fb5533e4205d3c5", "parents": [ "d977772e17ccaa1903b2af736f6405ab3a9f05cc" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Jan 13 10:26:05 2026 -0500" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Tue Jan 13 07:38:07 2026 -0800" }, "message": "internal/http3: add basic request stream handling for the server\n\nAdd support for the server to handle basic request streams.\nSpecifically, reading HTTP headers \u0026 body in requests, and writing HTTP\nheaders \u0026 body in responses.\n\nMore sophisticated behaviors, such as dealing with 1xx status and\ntrailing headers, will be done in the future.\n\nFor golang/go#70914\n\nChange-Id: I6a00435d72a118efe2ab76e4c7c53b9dca1d63f6\nReviewed-on: https://go-review.googlesource.com/c/net/+/735820\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nAuto-Submit: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "d977772e17ccaa1903b2af736f6405ab3a9f05cc", "tree": "607647d4778a25f55cf4a350a809da3ddd3dd005", "parents": [ "eea413e2942fbb59b323a2af0b1740da4d8aa93e" ], "author": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Jan 12 08:07:41 2026 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Jan 12 08:15:11 2026 -0800" }, "message": "go.mod: update golang.org/x dependencies\n\nUpdate golang.org/x dependencies to their latest tagged versions.\n\nChange-Id: I335bdc9150b51cb7d8ebd29721580dcdd5cb49ba\nReviewed-on: https://go-review.googlesource.com/c/net/+/735680\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Gopher Robot \u003cgobot@golang.org\u003e\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\nReviewed-by: Cherry Mui \u003ccherryyz@google.com\u003e\n" }, { "commit": "eea413e2942fbb59b323a2af0b1740da4d8aa93e", "tree": "d9e2cb6d08d4dd180538c29907d96367455fd79a", "parents": [ "9ace223794aa203b4c877d08a1f7bf2f595f6242" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Thu Jan 08 14:57:52 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Jan 09 07:13:59 2026 -0800" }, "message": "internal/http3: use go1.25 synctest.Test instead of go1.24 synctest.Run\n\ninternal/http3 was written back in go1.24, and relies on synctest.Run\nand goexperiment.synctest. This prevents us from running tests with\nnewer versions of Go.\n\nThis CL updates the existing build constraint and synctest usages to\ngo1.25 so we can still run tests using go1.25 and tip. Support for\nrunning synctest with go1.24 is not kept, since go1.26 release (and\ntherefore x/net go.mod updating to go1.25) is expected soon.\n\nChange-Id: Iebfa82ebd1da4a06ba613ce0f45051f4691037fc\nReviewed-on: https://go-review.googlesource.com/c/net/+/734940\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "9ace223794aa203b4c877d08a1f7bf2f595f6242", "tree": "a4177d021ce62a60ff13715e0f15bfbed29d2160", "parents": [ "7d3dbb06ceb45c3180f4f446cd635e6b59a0b9c2" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Tue Jan 06 16:45:22 2026 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Tue Jan 06 15:45:35 2026 -0800" }, "message": "websocket: add missing call to resp.Body.Close\n\nWithin hybiClientHandshake, after getting a response from\nhttp.ReadResponse, the body is never closed.\n\nWhen running a debugger with all of our test cases, this seems to not\nmatter usually since the body is typically a http.noBody, whose Close\njust returns nil. However, this is not always the case. Therefore, this\nCL adds the missing resp.Body.Close call.\n\nFixes golang/go#76952\n\nChange-Id: I292d6ccc8eb101b806738aa7fc3e5446b623d861\nReviewed-on: https://go-review.googlesource.com/c/net/+/734240\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "7d3dbb06ceb45c3180f4f446cd635e6b59a0b9c2", "tree": "6912bedcdd249326e8be8daf08b506ebcf0a3237", "parents": [ "35e1306bddd863f360fb94480c5fed84229953f0" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Mon Dec 08 17:32:23 2025 -0500" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Mon Dec 08 15:11:37 2025 -0800" }, "message": "http2: buffer the most recently received PRIORITY_UPDATE frame\n\nPer RFC 9218, servers should buffer the most recently received\nPRIORITY_UPDATE frame. This CL implements said buffering within the RFC\n9218 priority write scheduler.\n\nFor golang/go#75500\n\nChange-Id: I259f4f6787053de6388ec513086cfa1b294fa607\nReviewed-on: https://go-review.googlesource.com/c/net/+/728401\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "35e1306bddd863f360fb94480c5fed84229953f0", "tree": "1d1a8a0b1307eea21a10e4e47add104b9868534d", "parents": [ "7c360367ab7e57c0cfb7aef368fc6acefaaac3b1" ], "author": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Dec 08 11:56:54 2025 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Dec 08 13:41:29 2025 -0800" }, "message": "go.mod: update golang.org/x dependencies\n\nUpdate golang.org/x dependencies to their latest tagged versions.\n\nChange-Id: I9679bad7bfc127003a80fc41dab3cf34aaff78be\nReviewed-on: https://go-review.googlesource.com/c/net/+/728182\nAuto-Submit: Gopher Robot \u003cgobot@golang.org\u003e\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\nReviewed-by: David Chase \u003cdrchase@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "7c360367ab7e57c0cfb7aef368fc6acefaaac3b1", "tree": "df93c6060b1fa65817afd6be0e88acef179de2c5", "parents": [ "ec11eccf5a0f725281df0cdf40bb7ebef51d57ea" ], "author": { "name": "Dmitri Shuralyov", "email": "dmitshur@golang.org", "time": "Fri Nov 28 12:06:56 2025 -0500" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Fri Nov 28 14:06:04 2025 -0800" }, "message": "http2, webdav, websocket: fix %q verb uses with wrong type\n\nCaught early by the improved vet check gated behind the 1.26 language\nversion combined with a tiplang builder that tests with 1.26 language\nversion.\n\nFixes golang/go#76574.\nFixes golang/go#76599.\nFixes golang/go#76547.\n\nChange-Id: If8e2266013df0a39fc980a1e9287f8cb75949811\nCq-Include-Trybots: luci.golang.try:x_net-gotip-linux-amd64-tiplang\nReviewed-on: https://go-review.googlesource.com/c/net/+/725220\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\nCommit-Queue: Alan Donovan \u003cadonovan@google.com\u003e\nReviewed-by: Alan Donovan \u003cadonovan@google.com\u003e\nAuto-Submit: Dmitri Shuralyov \u003cdmitshur@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Alan Donovan \u003cadonovan@google.com\u003e\n" }, { "commit": "ec11eccf5a0f725281df0cdf40bb7ebef51d57ea", "tree": "606806130d9e0305849c0eacfcdd4ea72448bd74", "parents": [ "bff14c52567061031b9761881907c39e24792736" ], "author": { "name": "Marcus Efraimsson", "email": "marcus.efraimsson@gmail.com", "time": "Sat Nov 22 00:01:52 2025 +0100" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Nov 24 17:04:07 2025 -0800" }, "message": "trace: fix data race in RenderEvents\n\nThis commit updates the RenderEvents function by moving the read\nlock for event families so that it happens before checking the\nlength of event families and by that fixing a data race problem.\n\nFixes golang/go#76402\n\nChange-Id: Id02ebaa25cec8ef631aaf928ba084b0692e84583\nReviewed-on: https://go-review.googlesource.com/c/net/+/723160\nAuto-Submit: Sean Liao \u003csean@liao.dev\u003e\nReviewed-by: Sean Liao \u003csean@liao.dev\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "bff14c52567061031b9761881907c39e24792736", "tree": "4332056fd5a416940b844cad4c4a1350a72c7630", "parents": [ "88a642172c174ab11f4c56f0ede777de3c8a21d4" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Nov 13 09:52:54 2025 -0800" }, "committer": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Nov 24 14:35:53 2025 -0800" }, "message": "http2: don\u0027t PING a responsive server when resetting a stream\n\nWhen sending a RST_STREAM for a canceled request, we sometimes send\na PING frame along with the reset to confirm that the server is responsive\nand has received the reset.\n\nSending too many PINGs trips denial-of-service detection on some servers,\ncausing them to close a connection with an ENHANCE_YOUR_CALM error.\n\nDo not send a PING frame along with an RST_STREAM if the connection\nhas displayed signs of life since the canceled request began.\nSpecifically, if we\u0027ve received any stream-related frames since the\nrequest was sent, assume the server is responsive and do not send a PING.\n\nWe still send a PING if a request is canceled and no stream-related\nframes have been received from the server since the request was first\nsent.\n\nFor golang/go#76296\n\nChange-Id: I1be3532febf9ac99d65e9cd35346c02306db5f9d\nReviewed-on: https://go-review.googlesource.com/c/net/+/720300\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "88a642172c174ab11f4c56f0ede777de3c8a21d4", "tree": "61eaa166c5626f86605d053649092391302e49c3", "parents": [ "123d099e1bd872b38247bbcf9856540b8420d18d" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Nov 24 14:25:48 2025 -0800" }, "committer": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Nov 24 14:33:11 2025 -0800" }, "message": "dns/dnsmessage: avoid use of \"strings\" and \"math\" in dns/dnsmessage\n\ndns/dnsmessage is used by the net package, which does not have\nstrings and math in its permitted dependencies.\n\nChange-Id: I2b59887c59891dc61e49dd6430f3a72486ddd66a\nReviewed-on: https://go-review.googlesource.com/c/net/+/723902\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "123d099e1bd872b38247bbcf9856540b8420d18d", "tree": "7af4062feef27a573d1b404230954f952e60b7ac", "parents": [ "346cc6157ee53301dea14e57a45c22368ab46e55" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Tue Nov 18 10:57:31 2025 -0800" }, "committer": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Mon Nov 24 11:30:53 2025 -0800" }, "message": "http2: support net/http.Transport.NewClientConn\n\nPermit net/http to create new HTTP/2 client connections.\n\nWe do this by adding a NewClientConn method to the type the HTTP/2 client\nregisters with net/http.Transport.RegisterProtocol, which creates a\npersistent connection from a net.Conn.\n\nNo tests in this CL. Tests will be in net/http, and will cover\nboth the HTTP/1 and HTTP/2 paths for NewClientConn.\n\nFor golang/go#75772\n\nChange-Id: Ib1a06b4d13fdd6008e5db9a090c6e9632029a2a4\nReviewed-on: https://go-review.googlesource.com/c/net/+/722200\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\n" }, { "commit": "346cc6157ee53301dea14e57a45c22368ab46e55", "tree": "e7448765eeb7c42e6a7c257f1c4da012a63ff0a1", "parents": [ "9a296438e54dff851a45667aa645a97003b44db5" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Fri Nov 21 14:41:28 2025 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Fri Nov 21 16:46:41 2025 -0800" }, "message": "webdav: relax test to check for any redirect status, not just 301\n\nCL 720820 changed net/http to use a 307 Temporary Redirect instead of\na 301 Moved Permanently when performing an automatic redirect under\nsome circumstances. Update tests in the webdav package to be agnostic\non the exact redirect status code.\n\nChange-Id: I71784a738d18928a98387ddbd5475d50b19d15bf\nReviewed-on: https://go-review.googlesource.com/c/net/+/723120\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Sean Liao \u003csean@liao.dev\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nAuto-Submit: Nicholas Husin \u003cnsh@golang.org\u003e\n" }, { "commit": "9a296438e54dff851a45667aa645a97003b44db5", "tree": "80ccac9ca07c0b1a36ab08b4093cdf5848b8e8cc", "parents": [ "07cefd8a6bb170785052142a96034f2b2f7115bc" ], "author": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Tue Nov 11 10:34:45 2025 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Tue Nov 11 10:55:31 2025 -0800" }, "message": "go.mod: update golang.org/x dependencies\n\nUpdate golang.org/x dependencies to their latest tagged versions.\n\nChange-Id: I166e970128861674939fc46eaad37f74d19f1745\nReviewed-on: https://go-review.googlesource.com/c/net/+/719680\nReviewed-by: David Chase \u003cdrchase@google.com\u003e\nReviewed-by: Dmitri Shuralyov \u003cdmitshur@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Gopher Robot \u003cgobot@golang.org\u003e\n" }, { "commit": "07cefd8a6bb170785052142a96034f2b2f7115bc", "tree": "89bf0ed04f89ba3c29386a13c500df0636a45e40", "parents": [ "5ac9daca088ab4f378d7df849f6c7d28bea86071" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Sat Nov 08 11:48:50 2025 -0800" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Sat Nov 08 12:28:41 2025 -0800" }, "message": "context: deprecate\n\nLong superseded by the stdlib\u0027s context package.\n\nFixes golang/go#49506\n\nChange-Id: I16e9a1275a7370cd78d11180f608ae28ecdce143\nReviewed-on: https://go-review.googlesource.com/c/net/+/719060\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Alan Donovan \u003cadonovan@google.com\u003e\n" }, { "commit": "5ac9daca088ab4f378d7df849f6c7d28bea86071", "tree": "45ae01a480bc1318fcf4dd1cf183b014b6cae3a0", "parents": [ "d1f64cc67036f272ea900194674afcf71eb3e816" ], "author": { "name": "Sean Liao", "email": "sean@liao.dev", "time": "Sun Oct 26 22:34:45 2025 +0000" }, "committer": { "name": "Sean Liao", "email": "sean@liao.dev", "time": "Tue Nov 04 13:39:14 2025 -0800" }, "message": "publicsuffix: don\u0027t treat ip addresses as domain names\n\nWhile IP addresses are not domain names and probably\nshouldn\u0027t be passed to these functions at all, it seems\nwrong to have it handle IPv4 and IPv6 differently.\n\nFixes golang/go#32979\n\nChange-Id: Id321a08b552c11d990c3966636b64793f762143f\nReviewed-on: https://go-review.googlesource.com/c/net/+/715100\nReviewed-by: Michael Pratt \u003cmpratt@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "d1f64cc67036f272ea900194674afcf71eb3e816", "tree": "38d4d5c6dcad60571107654c6db5bdce41eb85df", "parents": [ "fff0469cf57f71bea0cbc4cc8cbfb8f92be7a417" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Oct 23 15:17:56 2025 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Mon Oct 27 13:48:05 2025 -0700" }, "message": "quic: use testing/synctest\n\nReplace bespoke fake time and synchronization with testing/synctest.\n\nChange-Id: Ic3fe9635dbad36c890783c38e00708c6cb7a15f8\nReviewed-on: https://go-review.googlesource.com/c/net/+/714482\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "fff0469cf57f71bea0cbc4cc8cbfb8f92be7a417", "tree": "8eb39c7831c0728a0e8ecb311f9632e31ed47d57", "parents": [ "f35e3a4dd2a34d1d47fac5b66210a40e2d374a79" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Oct 24 16:20:02 2025 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Oct 24 15:04:41 2025 -0700" }, "message": "http2: document that RFC 7540 prioritization does not work with small payloads\n\nThis change demonstrates that golang/go#75936 applies to the RFC 7540\nwrite scheduler.\n\nA similar test will be added for RFC 9218 write scheduler after support\nfor it is incorporated within http2/server.go.\n\nFor golang/go#75936\n\nChange-Id: I4e05dbeb0aab71942eb699b67383ef5b52c3ef4d\nReviewed-on: https://go-review.googlesource.com/c/net/+/714741\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\n" }, { "commit": "f35e3a4dd2a34d1d47fac5b66210a40e2d374a79", "tree": "4289d0e2c4f1c976654431bed7f6b1b6823cd316", "parents": [ "89adc90ac46a688eeda3a4a95b71acf073bd059c" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Oct 24 16:26:54 2025 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Oct 24 15:04:30 2025 -0700" }, "message": "http2: fix weight overflow in RFC 7540 write scheduler\n\nWe use uint8 (0-255, inclusive) to represent the RFC 7540 priorities\nweight (1-256, inclusive). To account for the difference, we add 1 to\nthe uint8 weight value within sortPriorityNodeSiblingsRFC7540.\n\nHowever, the addition was done before converting the uint8 type to\nfloat. As a result, when provided a maximum weight value, overflow will\nhappen and will cause the scheduler to treat the maximum weight as a\nminimum weight instead.\n\nThis CL fixes the issue by making sure the addition happens after the\ntype conversion.\n\nChange-Id: I404e87e5ad85fa06d5fa49cda613c93ac8847bdc\nReviewed-on: https://go-review.googlesource.com/c/net/+/714742\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\n" }, { "commit": "89adc90ac46a688eeda3a4a95b71acf073bd059c", "tree": "e18da7f34afc970eca7e9e6dad4e5d73fb0b98a0", "parents": [ "8d76a2c082f6f995c2ef1ce8a0d56f2e33b48092" ], "author": { "name": "Nicholas S. Husin", "email": "nsh@golang.org", "time": "Fri Oct 24 15:24:52 2025 -0400" }, "committer": { "name": "Nicholas Husin", "email": "nsh@golang.org", "time": "Fri Oct 24 15:04:23 2025 -0700" }, "message": "http2: fix typo referring to RFC 9218 as RFC 9128 instead\n\nIn go.dev/cl/704758, I accidentally typed RFC 9128 rather than 9218.\nThis changes fixes all such typo.\n\nFor go/golang#75500\n\nChange-Id: I61c3008a85ca0ca0e9f58315cb5eb20cd82fb9f9\nReviewed-on: https://go-review.googlesource.com/c/net/+/714740\nReviewed-by: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\n" }, { "commit": "8d76a2c082f6f995c2ef1ce8a0d56f2e33b48092", "tree": "a5a82c6731a80112251134582d649b34811cb876", "parents": [ "027f8b700eb72af5c0175a945535a42670628351" ], "author": { "name": "Damien Neil", "email": "dneil@google.com", "time": "Thu Oct 23 15:44:43 2025 -0700" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Fri Oct 24 07:51:19 2025 -0700" }, "message": "quic: don\u0027t defer MAX_STREAMS frames indefinitely\n\nAvoid a state where we can have a MAX_STREAMS frame to send,\nbut do not send the frame for an indefinite amount of time.\n\nConn.appendStreamFrames writes stream-related frames to\nthe current packet. It also handles removing streams\nfrom the Conn when we no longer need to track their state.\n\nRemoving streams can affect the frames we want to send.\nIn particular, we may want to send a MAX_STREAMS to the\npeer indicating that it can open more streams because\nwe\u0027ve closed out some of the existing ones.\n\nAdd MAX_STREAMS after removing streams, to ensure we\npick up any changes to the sent value before adding it.\n\nThis case doesn\u0027t show up in tests, because the test harness\u0027s\nidleness detection causes the Conn\u0027s event loop to run and notice\nthe pending MAX_STREAMS frame. Changing tests to use\ntesting/synctest (a followup CL) causes the problem to\nappear, because the event loop isn\u0027t run while the Conn\nis idle.\n\nChange-Id: Ia7394891317dae6ecfd529a9b3501ac082cb453e\nReviewed-on: https://go-review.googlesource.com/c/net/+/714481\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nAuto-Submit: Damien Neil \u003cdneil@google.com\u003e\nReviewed-by: Nicholas Husin \u003chusin@google.com\u003e\nReviewed-by: Nicholas Husin \u003cnsh@golang.org\u003e\n" } ], "next": "027f8b700eb72af5c0175a945535a42670628351" }