{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":56263501,"defaultBranch":"main","name":"ChezScheme","ownerLogin":"cisco","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2016-04-14T19:10:25.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/1376999?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1724168454.0","currentOid":""},"activityList":{"items":[{"before":"a982d6b0ea9bd520d7287be4299dac8f7c336ff2","after":"f01e22d6c57e5072d7c11fd9bfd2f2954f61d674","ref":"refs/heads/main","pushedAt":"2024-09-09T13:53:31.000Z","pushType":"pr_merge","commitsCount":5,"pusher":{"login":"jltaylor-us","name":"Jamie Taylor","path":"/jltaylor-us","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/18684406?s=80&v=4"},"commit":{"message":"doc,release_notes: clarify that ios section refers to native code","shortMessageHtmlLink":"doc,release_notes: clarify that ios section refers to native code"}},{"before":"74ca188c971bce295b4cf77a19c98557ba043dd4","after":"a982d6b0ea9bd520d7287be4299dac8f7c336ff2","ref":"refs/heads/main","pushedAt":"2024-08-28T13:46:31.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"Chez Scheme: improve support for lists in cptypes (#858)\n\nMove `list` form the normalptr slot to the multiplet slot.\r\nIn particular, this allows the reduction of\r\n\r\n (lambda (x f)\r\n (unless (list-assuming-immutable? x)\r\n (f)\r\n (list-assuming-immutable? x)))\r\n\r\n`pair`s are split in `list-pairs` and `nonlist-pairs`.\r\nA `pair` may go from one classification to the other,\r\nso the internal representations of the `list?` predicate\r\nuse both of them or neither.\r\n\r\nCo-authored-by: Gustavo Massaccesi ","shortMessageHtmlLink":"Chez Scheme: improve support for lists in cptypes (#858)"}},{"before":"23b2a1c9e6c8e0222cb9dbc9d73d3ca822d75912","after":null,"ref":"refs/heads/bburger/flvector","pushedAt":"2024-08-20T15:40:54.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"}},{"before":"3d89d4e2ae1e3c45d7b80366016ab775cc417df1","after":"74ca188c971bce295b4cf77a19c98557ba043dd4","ref":"refs/heads/main","pushedAt":"2024-08-20T15:40:51.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"},"commit":{"message":"report \"non-flonum found in flvector\", not non-fixnum (#865)","shortMessageHtmlLink":"report \"non-flonum found in flvector\", not non-fixnum (#865)"}},{"before":"94c057985aa29a3aabc6e6ca771fbfdaf813daa3","after":"23b2a1c9e6c8e0222cb9dbc9d73d3ca822d75912","ref":"refs/heads/bburger/flvector","pushedAt":"2024-08-20T14:51:24.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"},"commit":{"message":"report \"non-flonum found in flvector\", not non-fixnum","shortMessageHtmlLink":"report \"non-flonum found in flvector\", not non-fixnum"}},{"before":null,"after":"94c057985aa29a3aabc6e6ca771fbfdaf813daa3","ref":"refs/heads/bburger/flvector","pushedAt":"2024-08-20T14:31:35.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"},"commit":{"message":"report \"non-flonum found in flvector\", not non-fixnum","shortMessageHtmlLink":"report \"non-flonum found in flvector\", not non-fixnum"}},{"before":"f0be891d24a82c9282130d889c6394172f824eb7","after":"3d89d4e2ae1e3c45d7b80366016ab775cc417df1","ref":"refs/heads/main","pushedAt":"2024-08-16T14:57:38.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"makefiles/boot.zuo: remove broken main submodule (#862)\n\nThe main submodule hasn't worked since at least commit 10d59cc, and\r\ndocumented build modes have been added in the mean time that support\r\nits former uses.\r\n\r\nCloses https://github.com/cisco/ChezScheme/issues/810","shortMessageHtmlLink":"makefiles/boot.zuo: remove broken main submodule (#862)"}},{"before":"682265132ddb92f6fbf7a0bd9663658d538c4200","after":"f0be891d24a82c9282130d889c6394172f824eb7","ref":"refs/heads/main","pushedAt":"2024-08-06T19:43:11.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"repair check in `$describe-fasl-from-port` (#859)\n\nThis is an internal function, but it's used by a decompilation tool\r\nthat's included with Racket: `raco decompile --partial-fasl`.","shortMessageHtmlLink":"repair check in $describe-fasl-from-port (#859)"}},{"before":"2b1969a536da2d36dcd0f38ac67fab8feb0daa33","after":"682265132ddb92f6fbf7a0bd9663658d538c4200","ref":"refs/heads/main","pushedAt":"2024-08-06T19:42:55.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"fix foreign test for i3nt (#860)\n\nFor some build environments, `_WIN32` is needed in \"foreign3.c\" (as in\r\nother \"foreign*.c\" files) instead of `WIN32` without the leading\r\nunderscore. Also, simplify \"mats/build.zuo\" by using `_WIN32` in\r\n\"ftype.h\".","shortMessageHtmlLink":"fix foreign test for i3nt (#860)"}},{"before":"aab91cfd117c5da9284643c409b9561963d0e348","after":"2b1969a536da2d36dcd0f38ac67fab8feb0daa33","ref":"refs/heads/main","pushedAt":"2024-07-31T22:03:16.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"small error-message consistency improvement for bit indices (#852)\n\nFunctions like `vector-ref`, `string-ref`, `substring`, and\r\n`bitwise-bit-field` report a non-exact-integer index as a \"bad index\"\r\nerror. But `bitwise-bit-set?`, `logbit?`, `logbit0`, and `logbit1`\r\nsaid \"not an exact integer\", instead, while reporting \"bad index\" for\r\na negative exact integer; `bitwise-copy-bit` said \"not a nonnegative\r\nexact integer\". This commit changes the error message for those last\r\nfive functions to always say \"bad index\".\r\n\r\nFixnum variants of the operations can still say \"not a fixnum\" or\r\n\"invalid bit index\", depending on whether the index is a negative\r\nfixnum. That seems justified on the grounds that fixnum operations add\r\nan extra constraint on their arguments. Another possible direction is\r\nto change the error messages to consistently \"not a nonnegative exact\r\ninteger\" and \"not a nonnegative fixnum\", since (unlike other index\r\ncases), the index validity is independent of the indexed value. After\r\ntrying out both, sticking with \"invalid bit index\" seems clearest and\r\nthe most in the spirit of the current error messages.","shortMessageHtmlLink":"small error-message consistency improvement for bit indices (#852)"}},{"before":"0085c9fb66cf27f49fe252d6318a1d7b46752aa0","after":"aab91cfd117c5da9284643c409b9561963d0e348","ref":"refs/heads/main","pushedAt":"2024-07-29T03:54:27.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"configure: automatically detect endianness and word size on Linux (#855)","shortMessageHtmlLink":"configure: automatically detect endianness and word size on Linux (#855)"}},{"before":"55130b6d8788930ad428e6a20a43ebd0bb763c06","after":"0085c9fb66cf27f49fe252d6318a1d7b46752aa0","ref":"refs/heads/main","pushedAt":"2024-07-29T03:24:38.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"fix missing `wcwidth` declaration on GNU Hurd (#853)","shortMessageHtmlLink":"fix missing wcwidth declaration on GNU Hurd (#853)"}},{"before":"9e6afe33d8f416045462a6d8e9473bf98007057f","after":"55130b6d8788930ad428e6a20a43ebd0bb763c06","ref":"refs/heads/main","pushedAt":"2024-07-29T02:53:25.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"fix bytecode endianness handling on s390x, m68k and hppa (#854)","shortMessageHtmlLink":"fix bytecode endianness handling on s390x, m68k and hppa (#854)"}},{"before":"1524f8065d8e731f6f8be2caaf36d296f4b91a32","after":"9e6afe33d8f416045462a6d8e9473bf98007057f","ref":"refs/heads/main","pushedAt":"2024-07-29T02:51:23.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"fix the whatis lines in scheme.1.in (#856)","shortMessageHtmlLink":"fix the whatis lines in scheme.1.in (#856)"}},{"before":"f1ad314a3809074c6f2b986a5db6a84ad88700c1","after":"1524f8065d8e731f6f8be2caaf36d296f4b91a32","ref":"refs/heads/main","pushedAt":"2024-07-17T08:03:52.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"improve cross-compile instructions and makefile support (#849)\n\nAdd a `--cross` argument to `configure` to indicate that the generated\r\nmakefile's `make` should be like `make kernel`, and a pb-based\r\nbootstrap should rerun `configure` to build for the host machine\r\ninstead of the specified target machine.\r\n\r\nAlso, add support for `DESTDIR` as a more conventional alternative to\r\nthe `--temproot` configure flag. While we're at it, recognize\r\n`CFLAGS_FOR_BUILD` for building `bin/zuo`.\r\n\r\nAs part of the change to `configure`, having `CC_FOR_BUILD` set does\r\nnot by itself disable the use of pb for creating or updating boot\r\nfiles. Depending on how boot files are obtained, `--force` now may be\r\nneeded when using `CC_FOR_BUILD` as part of a cross compile that\r\ndoesn't use `--cross`.","shortMessageHtmlLink":"improve cross-compile instructions and makefile support (#849)"}},{"before":"3d1579e6c67e145895e6a0e7d0f9bf2b8853fbb3","after":"f1ad314a3809074c6f2b986a5db6a84ad88700c1","ref":"refs/heads/main","pushedAt":"2024-06-25T14:10:16.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"use `__has_builtin` to detect available intrinsics (#846)","shortMessageHtmlLink":"use __has_builtin to detect available intrinsics (#846)"}},{"before":"089a481b15a4cb245cd8647d95477803292631d0","after":"3d1579e6c67e145895e6a0e7d0f9bf2b8853fbb3","ref":"refs/heads/main","pushedAt":"2024-06-17T23:24:35.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"},"commit":{"message":"make \"V\" format directive case insensitive (#843)","shortMessageHtmlLink":"make \"V\" format directive case insensitive (#843)"}},{"before":"b06ba65dfb69a5a65969ea44c1459dd67e1a1c84","after":"089a481b15a4cb245cd8647d95477803292631d0","ref":"refs/heads/main","pushedAt":"2024-06-13T14:56:13.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"Allow building against cosmopolitan libc (#837)","shortMessageHtmlLink":"Allow building against cosmopolitan libc (#837)"}},{"before":"5396ce9e32f2bfaf0ce1a5f5baf53ad78745e162","after":null,"ref":"refs/heads/csug10","pushedAt":"2024-06-11T12:09:53.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"}},{"before":"fa451a1f6c356959290f2c60d2102c95d07bacb8","after":"b06ba65dfb69a5a65969ea44c1459dd67e1a1c84","ref":"refs/heads/main","pushedAt":"2024-06-11T12:09:50.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"},"commit":{"message":"Update references from csug9.5 to csug10.0 (#839)\n\n* update references from csug9.5 to csug10.0\r\n* update release notes folder","shortMessageHtmlLink":"Update references from csug9.5 to csug10.0 (#839)"}},{"before":"179fd95850cfca35d767ba6c09b93b3a0cf39c99","after":"5396ce9e32f2bfaf0ce1a5f5baf53ad78745e162","ref":"refs/heads/csug10","pushedAt":"2024-06-10T18:00:32.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"},"commit":{"message":"update release notes folder","shortMessageHtmlLink":"update release notes folder"}},{"before":null,"after":"179fd95850cfca35d767ba6c09b93b3a0cf39c99","ref":"refs/heads/csug10","pushedAt":"2024-06-10T17:56:54.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"burgerrg","name":"Bob Burger","path":"/burgerrg","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6304842?s=80&v=4"},"commit":{"message":"update references from csug9.5 to csug10.0","shortMessageHtmlLink":"update references from csug9.5 to csug10.0"}},{"before":"66c40f11423cf0718a0676b3cb09a0abd4990ec1","after":"fa451a1f6c356959290f2c60d2102c95d07bacb8","ref":"refs/heads/main","pushedAt":"2024-05-25T11:38:08.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"restore `(debug-level 2)` behavior, but reduce its guarantees (#834)\n\nWhen `debug-level` is 2 or more, optimization is supposed to refrain\r\nfrom moving a call to an error function into tail position of the\r\nenclosing function. (That's a stronger guarantee than the usual one of\r\nnot moving an expression into tail position if there's potentially a\r\nway to detect the movement through continuation marks.) An earlier\r\ncommit 6a73b9edc7 strengthened movement of error calls in a way that\r\ndid not preserve this `(debug-level 2)` constraint, so this commit\r\nprimarily adjusts those changes. Also, the earlier commit 19af32e2d\r\nallowed the compiler to move an error call out of tail position, and\r\nthat transformation is now suppressed when `debug-level` is 2 or more.\r\n\r\nThese changes are meant to help with debugging, but they're not enough\r\nto implement the previously specified behavior of `debug-level`. It\r\nturns out that Chez Scheme v9.x didn't implement the specified\r\nbehavior, either, because it would convert `(let ([x (error ...)]) x)`\r\nto just `(error ...)`, for example. The old specification seems too\r\nstrong, and so part of the revision here is to change the specificaton\r\nto be encouraging a particular interaction of errors and\r\ncontinuaton-inspection results, but not guaranteeing it. Meanwhile,\r\nwhen well-defined reflection on continuations is needed, continuation\r\nmarks provide that functionality.","shortMessageHtmlLink":"restore (debug-level 2) behavior, but reduce its guarantees (#834)"}},{"before":"ad064de67269f6dd06e1bd8a7ac4324ee2a6324f","after":"66c40f11423cf0718a0676b3cb09a0abd4990ec1","ref":"refs/heads/main","pushedAt":"2024-04-30T16:23:33.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"repair `pseudo-random-generator-next!` for large integers (#831)\n\nFor an argument exact integer greater than 4294967087, generation of a\r\nrandom integer by piecing together smaller random numbers was broken.\r\nIt didn't cover the whole range uniformly, and by mistreating the\r\nlargest part of the integer, it could always produce 0 (e.g., for\r\n`(expt 2 32)`).\r\n\r\nCloses #830","shortMessageHtmlLink":"repair pseudo-random-generator-next! for large integers (#831)"}},{"before":"08a45e564b33bf75df626479427d5fefbd8c921a","after":"ad064de67269f6dd06e1bd8a7ac4324ee2a6324f","ref":"refs/heads/main","pushedAt":"2024-04-26T14:28:09.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"configure: improve inference of compiler flags for pb builds (#829)\n\nWhen building an pb variant like `tpb64l`, and even when building for\r\na platform without native-code support, the \"configure\" script may\r\nstill be able to infer suitable C compilation and linking flags for\r\nthe kernel. Add an `--os=...` flag that can be used to override that\r\ninference in case it's not correct (as in the case of cross\r\ncompilation).","shortMessageHtmlLink":"configure: improve inference of compiler flags for pb builds (#829)"}},{"before":"c7a35602eb8a2115010a829fb7750d9fb2c8de5c","after":"08a45e564b33bf75df626479427d5fefbd8c921a","ref":"refs/heads/main","pushedAt":"2024-04-13T00:48:00.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"clean up memory fence and compare-and-swap kernel functions (#826)\n\nCAS variants with different fence properties in \"atomic.h\" ware a\r\nleftover from an early experiment in parallel garbage collection, and\r\nthat experiment did not turn out to be a good idea. The CAS variants\r\nwere left half-implemented, and they're not needed, so this commit\r\nsimplifies to reflect the needed functionality. Also, the commit adds\r\ncomments to clarify that CAS support is used only for pb mode and to\r\nexplain trade-offs for inline assembly and compiler intrinsics.\r\n\r\nSmall repairs to the build scripts enable pbarch builds on Windows,\r\nfor what it's worth. Otherwise, this is just an internal clean-up and\r\ndoes not add anything new or fix any bugs.","shortMessageHtmlLink":"clean up memory fence and compare-and-swap kernel functions (#826)"}},{"before":"66dd7a813934f71ec988d5b0ecd98b4b9386abb4","after":"c7a35602eb8a2115010a829fb7750d9fb2c8de5c","ref":"refs/heads/main","pushedAt":"2024-04-08T20:13:49.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"add missing memory fences for tppc32le and pb variants (#825)\n\nFor tppc32le, memory-fence generation needed to be turned on in\r\ncpprim. For pb variants, use `__sync_synchronize()` as provided by GCC\r\nand Clang to support architectures other than ones explicitly covered\r\nin \"atomic.h\".","shortMessageHtmlLink":"add missing memory fences for tppc32le and pb variants (#825)"}},{"before":"9576b83dd757cf1494933c9fbc80cb6aff022295","after":"66dd7a813934f71ec988d5b0ecd98b4b9386abb4","ref":"refs/heads/main","pushedAt":"2024-04-07T14:12:46.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"repairs for pb32 foreign interface (#824)\n\nSupport via libffi for pb32 had not been well tested before. Just\r\nrunning existing tests turned up several problems in the handling of\r\ncallable addresses and 64-bit arguments and results.","shortMessageHtmlLink":"repairs for pb32 foreign interface (#824)"}},{"before":"dbce392733569b6d910b66286c95d2fb2d6c91ff","after":"9576b83dd757cf1494933c9fbc80cb6aff022295","ref":"refs/heads/main","pushedAt":"2024-03-27T16:01:05.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"repair GC for reference bytevector whose length isn't a multiple of a word (#823)\n\nA reference bytevector is intended to be used as a multiple of a word\r\nsize, but there's no explicit constraint, and so it should work with\r\nany length. When the length is not a multiple of the word size, then\r\nin a copying step, extra bytes at the end of the bytevector were\r\nmissed.","shortMessageHtmlLink":"repair GC for reference bytevector whose length isn't a multiple of a…"}},{"before":"3555ad03063bb5f246b46e6239dc64cb01a36ee4","after":"dbce392733569b6d910b66286c95d2fb2d6c91ff","ref":"refs/heads/main","pushedAt":"2024-03-26T20:42:45.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"mflatt","name":"Matthew Flatt","path":"/mflatt","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/295034?s=80&v=4"},"commit":{"message":"reduce memory use of pbchunk conversion (#820)\n\nC code generated by the pbchunk conversion is especially verbose, with\r\nlong macro names for simple operations and lots of comments to reflect\r\nsource instructions. Generation of that code was formerly rendering to\r\ntext in memory, and then splitting into the output files, which turns\r\nout to use a lot of memory. Converting to text at the last minute can\r\nhave a much lower peak memory use.\r\n\r\nIf memory is at a premimum, then rendering to a temporary file would\r\nwork even better, but this change already brings memory use in line\r\nwith other build actions without having to deal with a temporary file.\r\nMemory use could also be reduced by dropping comments that connect to\r\nsource instructions; again, though, doesn't seem necessary for now.","shortMessageHtmlLink":"reduce memory use of pbchunk conversion (#820)"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEsPb7GAA","startCursor":null,"endCursor":null}},"title":"Activity · cisco/ChezScheme"}