repo stringclasses 15
values | fix_commit stringlengths 40 40 | buggy_commit stringlengths 40 40 | message stringlengths 3 64.3k | files listlengths 1 300 | timestamp timestamp[s]date 2013-03-13 20:45:00 2026-04-11 07:48:46 |
|---|---|---|---|---|---|
ollama/ollama | 33801c1597edca5dd04c5de117db89b5bc27f43a | e4340667e33e0efa5dee471917d71ad6011e59ba | Fixed Deepseek2 adding nil tensor error | [
{
"path": "model/models/deepseek2/model.go",
"patch": "@@ -150,7 +150,9 @@ func (moe *sparse) Moe(ctx ml.Context, hiddenStates, topKIndices, topKWeights ml\n }\n \n func (moe *sparse) topKIndices(ctx ml.Context, scores ml.Tensor, opts *Options) ml.Tensor {\n-\tscores = scores.Add(ctx, moe.ExpProbsBias)\n+\t... | 2025-10-03T21:20:06 |
huggingface/transformers | 187439c3fa139b2102a874483e9f8f0cfa8e5557 | ef976a7e181b78abf2f1ba7ea02e506ea1cb111e | VLM: special multimodal Tokenizer (#34461)
* kinda works
* update
* add tests
* update
* use special tokens in processors
* typo
* fix copies
* fix
* fix moshi after rebase
* update
* fix tests
* update
* Update docs/source/en/main_classes/tokenizer.md
Co-authored-by: Arthur <485... | [
{
"path": "docs/source/en/main_classes/tokenizer.md",
"patch": "@@ -51,6 +51,25 @@ token space (e.g., getting the index of the token comprising a given character o\n to a given token).\n \n \n+# Multimodal Tokenizer\n+\n+Apart from that each tokenizer can be a \"multimodal\" tokenizer which means that the t... | 2024-11-04T15:37:51 |
golang/go | a21de9ec73b8a433cafd336448dc8111a4e4571e | 8fa84772ba035b74975572fbc9df0330523cc388 | [dev.regabi] cmd/link: resolve symbol ABI in shared linkage
In shared build mode and linkage, currently we assume all
function symbols are ABI0 (except for generated type algorithm
functions), and alias them to ABIInternal. When the two ABIs
actually differ (as it is now), this is not actually correct.
This CL resolve... | [
{
"path": "src/cmd/link/internal/ld/lib.go",
"patch": "@@ -2091,6 +2091,26 @@ func ldshlibsyms(ctxt *Link, shlib string) {\n \t\tErrorf(nil, \"cannot read symbols from shared library: %s\", libpath)\n \t\treturn\n \t}\n+\n+\t// collect text symbol ABI versions.\n+\tsymabi := make(map[string]int) // map (unm... | 2021-02-03T20:07:33 |
nodejs/node | 11b6c0de4114fa499bcffbda91d8cea9d3bdea03 | a1a409a8ca2af0b12ee4b1397ef49e823e89327d | child_process: define EACCES as a runtime error
Access permission on the target child is currently thrown
as an exception. Bring this under the runtime error definition,
much like ENOENT and friends.
PR-URL: https://github.com/nodejs/node/pull/19294
Fixes: https://github.com/nodejs/help/issues/990
Reviewed-By: James ... | [
{
"path": "lib/internal/child_process.js",
"patch": "@@ -33,6 +33,7 @@ const { isUint8Array } = require('internal/util/types');\n const spawn_sync = process.binding('spawn_sync');\n \n const {\n+ UV_EACCES,\n UV_EAGAIN,\n UV_EINVAL,\n UV_EMFILE,\n@@ -315,7 +316,8 @@ ChildProcess.prototype.spawn = fun... | 2018-03-12T08:32:40 |
ollama/ollama | e4340667e33e0efa5dee471917d71ad6011e59ba | 2fa1e92a995e7599bf6403dc6055cf2f4a4d68ce | Workaround broken NVIDIA iGPU free VRAM data (#12490)
The CUDA APIs for reporting free VRAM are useless on NVIDIA iGPU
systems as they only return the kernels actual free memory and ignore
buff/cache allocations which on a typical system will quickly fill up
most of the free system memory. As a result, we incorrectly... | [
{
"path": "discover/runner.go",
"patch": "@@ -330,6 +330,9 @@ func GPUDevices(ctx context.Context, runners []FilteredRunnerDiscovery) []ml.Dev\n \t\t}\n \t}\n \n+\t// Apply any iGPU workarounds\n+\tiGPUWorkarounds(devices)\n+\n \treturn devices\n }\n \n@@ -540,3 +543,32 @@ func GetDevicesFromRunner(ctx cont... | 2025-10-03T19:17:21 |
rust-lang/rust | f894a81654d6fd67137ba0676bfbd295a48cb471 | 6ee4f34741877a4070c889f75dffe775f736ab36 | Fulfill expectations after first `missing-panics-doc` | [
{
"path": "clippy_lints/src/doc/missing_headers.rs",
"patch": "@@ -2,14 +2,13 @@ use super::{DocHeaders, MISSING_ERRORS_DOC, MISSING_PANICS_DOC, MISSING_SAFETY_D\n use clippy_utils::diagnostics::{span_lint, span_lint_and_note};\n use clippy_utils::macros::{is_panic, root_macro_call_first_node};\n use clippy... | 2025-03-22T16:25:19 |
huggingface/transformers | ef976a7e181b78abf2f1ba7ea02e506ea1cb111e | 33868a057c02f0368ba63bd1edb746be38fe3d90 | Update trainer for easier handling of accumulate, compile fixes, and proper reporting (#34511)
* Update trainer for easier handling of accumulate + proper reporting
* test
* Fixup tests
* Full fix
* Fix style
* rm comment
* Fix tests
* Minimize test + remove py 311 check
* Unused import
* Fo... | [
{
"path": "src/transformers/modeling_utils.py",
"patch": "@@ -28,7 +28,7 @@\n import warnings\n from contextlib import contextmanager\n from dataclasses import dataclass\n-from functools import lru_cache, partial, wraps\n+from functools import partial, wraps\n from threading import Thread\n from typing impo... | 2024-11-04T12:47:34 |
golang/go | b54cd94d478c95e79e5eea1d77e73d7b2b769f09 | 4516afebedd18692c6dc70cbdee16a049c26024b | embed, io/fs: clarify that leading and trailing slashes are disallowed
Fixes #44012
Change-Id: I5782cea301a65ae12ba870ff1e6b2e0a2651dc09
Reviewed-on: https://go-review.googlesource.com/c/go/+/290071
Run-TryBot: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <g... | [
{
"path": "src/embed/embed.go",
"patch": "@@ -61,12 +61,15 @@\n // The Go build system will recognize the directives and arrange for the declared variable\n // (in the example above, content) to be populated with the matching files from the file system.\n //\n-// The //go:embed directive accepts multiple sp... | 2021-02-05T21:45:40 |
nodejs/node | 855dabd675f3d1b06130bd1b92a131dca89f5eda | c6ae8a28109fc19ea30a5d785b2215345564c0bb | src: call CleanupHandles in FreeEnvironment
CleanupHandles() has not been called in our own code base anymore
after the v8 debug agent has been removed. It used to be in the
~Environment() destructor but then removed to avoid firing other
events after the exit event, given that we were not going to
clean up handles fo... | [
{
"path": "src/node.cc",
"patch": "@@ -4405,6 +4405,7 @@ Environment* CreateEnvironment(IsolateData* isolate_data,\n \n \n void FreeEnvironment(Environment* env) {\n+ env->CleanupHandles();\n delete env;\n }\n ",
"additions": 1,
"deletions": 0,
"language": "Unknown"
},
{
"path": "test... | 2018-03-13T13:07:10 |
rust-lang/rust | 6ee4f34741877a4070c889f75dffe775f736ab36 | c89368537cc76c83ecc7ea7282423f9a189e02aa | Abide by `allow`/`expect` in bodies for `missing_panics_doc` | [
{
"path": "clippy_lints/src/doc/missing_headers.rs",
"patch": "@@ -3,7 +3,7 @@ use clippy_utils::diagnostics::{span_lint, span_lint_and_note};\n use clippy_utils::macros::{is_panic, root_macro_call_first_node};\n use clippy_utils::ty::{get_type_diagnostic_name, implements_trait_with_env, is_type_diagnostic_... | 2025-03-14T15:22:33 |
ollama/ollama | 2fa1e92a995e7599bf6403dc6055cf2f4a4d68ce | 07e36761c380b838b2d4156dafde59105892b1ad | test: add template error test (#12489) | [
{
"path": "template/template_test.go",
"patch": "@@ -154,24 +154,55 @@ func TestTemplate(t *testing.T) {\n }\n \n func TestParse(t *testing.T) {\n-\tcases := []struct {\n+\tvalidCases := []struct {\n+\t\tname string\n \t\ttemplate string\n \t\tvars []string\n \t}{\n-\t\t{\"{{ .Prompt }}\", []string{... | 2025-10-03T19:05:34 |
golang/go | 4516afebedd18692c6dc70cbdee16a049c26024b | 8869086d8f0a31033ccdc103106c768dc17216b1 | testing/fstest: avoid symlink-induced failures in tester
Do not require directory entry and Stat result to match for symlinks,
because they won't (Stat dereferences the symlink).
Fixes #44113.
Change-Id: Ifc6dbce5719906e2f42254a7172f1ef787464a9e
Reviewed-on: https://go-review.googlesource.com/c/go/+/290009
Trust: Ru... | [
{
"path": "src/testing/fstest/testfs.go",
"patch": "@@ -403,22 +403,33 @@ func (t *fsTester) checkStat(path string, entry fs.DirEntry) {\n \t\treturn\n \t}\n \tfentry := formatEntry(entry)\n-\tfinfo := formatInfoEntry(info)\n-\tif fentry != finfo {\n-\t\tt.errorf(\"%s: mismatch:\\n\\tentry = %s\\n\\tfile.St... | 2021-02-05T15:55:12 |
vercel/next.js | ccc8d271dfe1a49999333c2fbd2a1b6105515183 | b16fcccfc069695c45252af566bc55c2cdfffb4d | Revert "edge-ssr: bundle next/dist as ESM for better tree-shaking (#40251) (#40967)
This reverts commit 11deaaa82bda301254167967903a384ef6c4c51f.
Temporarily reverts the above commit due to breaking middleware/edge
functions once deployed.
Fixes:
https://github.com/vercel/next.js/actions/runs/3133433920/jobs/5... | [
{
"path": "packages/next/amp.js",
"patch": "@@ -1,4 +1 @@\n-module.exports =\n- process.env.NEXT_RUNTIME === 'edge'\n- ? require('./dist/esm/shared/lib/amp')\n- : require('./dist/shared/lib/amp')\n+module.exports = require('./dist/shared/lib/amp')",
"additions": 1,
"deletions": 4,
"langua... | 2022-09-27T22:05:40 |
nodejs/node | d9d0a97541d7a9b478ecc500e5961ad349f42540 | 1ac14244763a7f87d51a517f0817eca82cbc86b6 | test: remove NODE_DEBUG in global module loading test
Otherwise the debug log output might be mixed up with
the expected errors and the assertion matching the error
message would fail.
PR-URL: https://github.com/nodejs/node/pull/19177
Refs: https://github.com/nodejs/node/pull/19112
Reviewed-By: Gus Caplan <me@gus.hos... | [
{
"path": "test/parallel/test-module-loading-globalpaths.js",
"patch": "@@ -41,8 +41,6 @@ if (process.argv[2] === 'child') {\n const testFixturesDir = fixtures.path(path.basename(__filename, '.js'));\n \n const env = Object.assign({}, process.env);\n- // Turn on module debug to aid diagnosing failures.... | 2018-03-14T12:56:28 |
huggingface/transformers | e2ac16b28a0b8b900e136750309ca40c49d975c5 | 86701f2b6ff2085a3cd3ad1d30bc2ff2b10fbd94 | Large modular logic refactoring (#34487)
* rework converter
* Update modular_model_converter.py
* Update modular_model_converter.py
* Update modular_model_converter.py
* Update modular_model_converter.py
* cleaning
* cleaning
* finalize imports
* imports
* Update modular_model_converter.py
... | [
{
"path": "examples/modular-transformers/configuration_my_new_model.py",
"patch": "@@ -1,9 +1,9 @@\n-# 🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨\n-# This file was automatically generated from <path_to_modular_file.py>.\n-# Do N... | 2024-11-01T09:13:51 |
rust-lang/rust | c89368537cc76c83ecc7ea7282423f9a189e02aa | ce39784f13022efa0f3b7f23c63150ae51c34d44 | Fix partially const bodies not linting `missing_panics_doc` | [
{
"path": "clippy_lints/src/doc/missing_headers.rs",
"patch": "@@ -1,7 +1,7 @@\n use super::{DocHeaders, MISSING_ERRORS_DOC, MISSING_PANICS_DOC, MISSING_SAFETY_DOC, UNNECESSARY_SAFETY_DOC};\n use clippy_utils::diagnostics::{span_lint, span_lint_and_note};\n use clippy_utils::macros::{is_panic, root_macro_ca... | 2025-03-14T15:02:00 |
ollama/ollama | c29fb007c0695210b833008775aed67a507f1f5a | 730ed6e9e12d0bf182d554a54dee8bbbef6a88c7 | CI: temporarily disable clang install (#12486)
This will likely yield builds that have problems with unicode characters
but at least we can start testing the release while we try to find an
alternate clang compiler for windows, or mingw ships a fixed version. | [
{
"path": ".github/workflows/release.yaml",
"patch": "@@ -186,7 +186,8 @@ jobs:\n if: matrix.arch == 'amd64'\n run: |\n $ErrorActionPreference = \"Stop\"\n- Start-Process \"C:\\msys64\\usr\\bin\\pacman.exe\" -ArgumentList @(\"-S\", \"--noconfirm\", \"mingw-w64-clang-x86_64... | 2025-10-03T03:31:18 |
golang/go | 370e9f58432c51bf3d95308cdc7109e25cc141f6 | ca2f15289337f57dcfb938000af249532f79e4d4 | [dev.typeparams] cmd/compile/internal/types2: use 512 bits as max. integer precision
This matches the compiler's existing limitations and thus ensures
that types2 reports the same errors for oversize integer constants.
Change-Id: I4fb7c83f3af69098d96f7b6c53dbe3eaf6ea9ee4
Reviewed-on: https://go-review.googlesource.co... | [
{
"path": "src/cmd/compile/internal/typecheck/const.go",
"patch": "@@ -449,7 +449,7 @@ func EvalConst(n ir.Node) ir.Node {\n \t\tn := n.(*ir.BinaryExpr)\n \t\tnl, nr := n.X, n.Y\n \t\tif nl.Op() == ir.OLITERAL && nr.Op() == ir.OLITERAL {\n-\t\t\t// shiftBound from go/types; \"so we can express smallestFloat... | 2021-01-29T23:29:36 |
vercel/next.js | b16fcccfc069695c45252af566bc55c2cdfffb4d | 0f41a48f8fd828c40527cab8cba7ad645d3290ce | remove legacy transform code (#40966)
* Remove legacy `__next_rsc__` transform
* Fix `staticGenerationAsyncStorage` is null during rendering | [
{
"path": "packages/next/build/webpack/loaders/next-flight-client-entry-loader.ts",
"patch": "@@ -20,25 +20,15 @@ export default async function transformSource(this: any): Promise<string> {\n }\n \n const requests = modules as string[]\n- const code =\n- requests\n- // Filter out css files on t... | 2022-09-27T22:03:49 |
rust-lang/rust | ce39784f13022efa0f3b7f23c63150ae51c34d44 | bb0d09b220dc94fa956d59b395c825b217a8070e | Move `FindPanicUnwrap` to `missing_headers.rs` | [
{
"path": "clippy_lints/src/doc/missing_headers.rs",
"patch": "@@ -1,9 +1,13 @@\n use super::{DocHeaders, MISSING_ERRORS_DOC, MISSING_PANICS_DOC, MISSING_SAFETY_DOC, UNNECESSARY_SAFETY_DOC};\n use clippy_utils::diagnostics::{span_lint, span_lint_and_note};\n+use clippy_utils::macros::{is_panic, root_macro_c... | 2025-03-14T14:47:03 |
huggingface/transformers | 86701f2b6ff2085a3cd3ad1d30bc2ff2b10fbd94 | 4cc0813e28c3ea1f0e1257d6079658c5f71d3dc6 | :red_circle: :red_circle: fix `query_pre_attn_scalar` different of `num_heads` in default gemma2 config (#34540)
* fix query_pre_attn_scalar different of num_heads in default config
* propagate modular changes
* fix copies
* fix modular copies
* fix copies?
* correct copies fix | [
{
"path": "src/transformers/models/gemma2/configuration_gemma2.py",
"patch": "@@ -36,15 +36,15 @@ class Gemma2Config(PretrainedConfig):\n vocab_size (`int`, *optional*, defaults to 256000):\n Vocabulary size of the Gemma2 model. Defines the number of different tokens that can be represen... | 2024-11-01T08:06:17 |
nodejs/node | e5e0f9f3a839e261faa6dc3e2c4fedcd46bef5f4 | 8d336dd8b1268f2b23f72999ebbd54f1c12263b8 | doc: fix minor issues in async_hooks.md
* easily -> easy
* was -> is
* add a missing comma
PR-URL: https://github.com/nodejs/node/pull/19313
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com> | [
{
"path": "doc/api/async_hooks.md",
"patch": "@@ -147,10 +147,10 @@ unintentional side effects.\n Because printing to the console is an asynchronous operation, `console.log()`\n will cause the AsyncHooks callbacks to be called. Using `console.log()` or\n similar asynchronous operations inside an AsyncHooks ... | 2018-03-13T04:04:55 |
ollama/ollama | 730ed6e9e12d0bf182d554a54dee8bbbef6a88c7 | dc06601677b6d5a833ae8206778b4acfdc290b81 | ci: fix windows build (#12485) | [
{
"path": ".github/workflows/release.yaml",
"patch": "@@ -163,7 +163,7 @@ jobs:\n cmake --preset \"${{ matrix.preset }}\" ${{ matrix.flags }} -DOLLAMA_RUNNER_DIR=\"${{ matrix.runner_dir }}\"\n cmake --build --parallel --preset \"${{ matrix.preset }}\"\n cmake --install build --... | 2025-10-03T02:16:01 |
golang/go | 120b819f45d1c109a1c2ef380edde9e826862a5c | 52d5cb2822966c00ce2ef97eb08bec4850d76fb2 | [dev.regabi] go/types: report error for invalid main function signature
This is a port of CL 279424, which didn't make it into master in time
for go1.16. Move it to dev.regabi so that it may be merged.
Notably, this port no longer removes the _InvalidInitSig error code,
instead opting to deprecate it. Now that error ... | [
{
"path": "src/go/types/decl.go",
"patch": "@@ -751,8 +751,12 @@ func (check *Checker) funcDecl(obj *Func, decl *declInfo) {\n \tobj.typ = sig // guard against cycles\n \tfdecl := decl.fdecl\n \tcheck.funcType(sig, fdecl.Recv, fdecl.Type)\n-\tif sig.recv == nil && obj.name == \"init\" && (sig.params.Len() >... | 2021-02-04T16:16:25 |
huggingface/transformers | 4cc0813e28c3ea1f0e1257d6079658c5f71d3dc6 | 6beb3f1691ca13556c7421b2c0503068906137a2 | BLIP: enable generation tests (#34174)
* blip2 tests
* instructblips
* copies
* fix slow tests
* fix
* uncomment this
* clean up after rebase
* should be model main input
* fix overwritten tests
* oops len should be multiple of frame number
* style
* fix some tests | [
{
"path": "src/transformers/models/blip_2/modeling_blip_2.py",
"patch": "@@ -2342,24 +2342,11 @@ def generate(\n )\n generate_kwargs[\"min_length\"] = generate_kwargs.get(\"min_length\", 0) + language_model_inputs.shape[1]\n \n- outputs = self.language_model.generate(\... | 2024-11-01T07:54:48 |
ollama/ollama | dc06601677b6d5a833ae8206778b4acfdc290b81 | 1ed2881ef05cd62d97f3fc3687301f9c69249e3b | ci: fix windows build (#12484) | [
{
"path": ".github/workflows/release.yaml",
"patch": "@@ -163,7 +163,7 @@ jobs:\n cmake --preset \"${{ matrix.preset }}\" ${{ matrix.flags }} -DOLLAMA_RUNNER_DIR=\"${{ matrix.runner_dir }}\"\n cmake --build --parallel --preset \"${{ matrix.preset }}\"\n cmake --install build --... | 2025-10-03T01:59:26 |
golang/go | 52d5cb2822966c00ce2ef97eb08bec4850d76fb2 | bc451b5770dc99b6a74934c26fd11a8cdc172bb1 | [dev.regabi] cmd/internal/obj: access Attribute atomically
Symbol's Attributes and ABI are in the same word. In the
concurrent backend, we may read one symbol's ABI (the callee)
while setting its attributes in another goroutine.
Fix racecompile build.
Change-Id: I500e869bafdd72080119ab243db94eee3afcf926
Reviewed-on:... | [
{
"path": "src/cmd/internal/obj/link.go",
"patch": "@@ -39,6 +39,7 @@ import (\n \t\"cmd/internal/sys\"\n \t\"fmt\"\n \t\"sync\"\n+\t\"sync/atomic\"\n )\n \n // An Addr is an argument to an instruction.\n@@ -647,37 +648,52 @@ const (\n \tattrABIBase\n )\n \n-func (a Attribute) DuplicateOK() bool { re... | 2021-02-03T23:10:04 |
vercel/next.js | e0cc9cd44ffd11fa7b446ffe7f25a35fca8f66f3 | bc685142aa14b9f2dcfe0208c11b3f8fbf0eb5f4 | feat(experimental): option to polyfill `fetch` using `undici` in Node.js <18 (#40318)
This PR adds a new `experimental.enableUndici` option to let the
developer switch from `next-fetch` to `undici` as the underlying
polyfill for `fetch` in Node.js.
In the current implementation, Next.js makes sure that `fetch` is... | [
{
"path": "jest.config.js",
"patch": "@@ -3,13 +3,17 @@ const nextJest = require('next/jest')\n const createJestConfig = nextJest()\n \n // Any custom config you want to pass to Jest\n+/** @type {import('jest').Config} */\n const customJestConfig = {\n testMatch: ['**/*.test.js', '**/*.test.ts', '**/*.tes... | 2022-09-27T20:37:28 |
huggingface/transformers | 6beb3f1691ca13556c7421b2c0503068906137a2 | b53e44e847230a39caecc320527776ed80a3ceff | Blip: get/set input embeddings correctly (#34152)
* set-get embeds
* add tests
* fix tests
* remove
* return dict True
* fix tests
* why did i remove this
* enabel torchscript tests | [
{
"path": "src/transformers/models/blip/modeling_blip.py",
"patch": "@@ -795,6 +795,12 @@ def __init__(self, config: BlipConfig):\n # Initialize weights and apply final processing\n self.post_init()\n \n+ def get_input_embeddings(self):\n+ return self.text_model.get_input_embedding... | 2024-11-01T07:39:39 |
nodejs/node | 90b05382734aca10b51b187eb955a964cbcaed74 | 52e869bee8f639fb2b256efa814446d0c325a4ac | test: use descriptive names for regression tests
Rename the tests appropriately alongside mentioning the subsystem.
Also, make a few basic changes to make sure the tests conform to the
standard test structure.
- Rename test-regress-GH-9819 to test-crypto-tostring-segfault
- Rename test-regress-GH-5051 to test-http-ad... | [
{
"path": "test/parallel/test-crypto-tostring-segfault.js",
"patch": "@@ -3,6 +3,11 @@ const common = require('../common');\n if (!common.hasCrypto)\n common.skip('missing crypto');\n \n+// This test ensures that node doesn't SEGFAULT when either of\n+// `crypto.createHash` or `crypto.createHmac` are give... | 2018-03-10T20:05:40 |
ollama/ollama | 1ed2881ef05cd62d97f3fc3687301f9c69249e3b | 0bda72892cc8b42a649e950f004b100020bcd1aa | templates: fix crash in improperly defined templates (#12483) | [
{
"path": "server/images.go",
"patch": "@@ -105,12 +105,16 @@ func (m *Model) Capabilities() []model.Capability {\n \n \tbuiltinParser := parsers.ParserForName(m.Config.Parser)\n \t// Check for tools capability\n-\tif slices.Contains(m.Template.Vars(), \"tools\") || (builtinParser != nil && builtinParser.Ha... | 2025-10-03T00:25:55 |
golang/go | afd67f333466fc67cd37433e45ecdb190efc8f51 | 401d7e5a242f1007c2637a25111a6fa985728c08 | [dev.regabi] go/types: no "declared but not used" errors for invalid var decls
This is a port of CL 274615, adapted to go/types. The only change was in
the positioning of expected errors in vardecl.src: in go/types they are
positioned on the identifier.
Change-Id: Iab03265a7c4287749373e4380c6db6a95f262f30
Reviewed-on... | [
{
"path": "src/go/types/assignments.go",
"patch": "@@ -120,6 +120,7 @@ func (check *Checker) initVar(lhs *Var, x *operand, context string) Type {\n \t\tif lhs.typ == nil {\n \t\t\tlhs.typ = Typ[Invalid]\n \t\t}\n+\t\tlhs.used = true\n \t\treturn nil\n \t}\n ",
"additions": 1,
"deletions": 0,
"la... | 2021-02-04T15:27:41 |
vercel/next.js | 377a39a88ddaf3f82353ebce656c02bbae251576 | a283b7b14ee93a28cde311c396cce071e61ef2d7 | Typo in middleware upgrade guide (#40963)
<!--
Thanks for opening a PR! Your contribution is much appreciated.
To make sure your PR is handled as smoothly as possible we request that
you follow the checklist sections below.
Choose the right checklist for the change that you're making:
-->
## Bug
- [ ] Relat... | [
{
"path": "errors/middleware-upgrade-guide.md",
"patch": "@@ -177,11 +177,11 @@ export default async function handler(req: NextRequest) {\n \n ### Summary of changes\n \n-| Added | Removed |\n-| ----------------------- | ------------- |\n-| `cookie.set` | `cookie` |\n... | 2022-09-27T19:40:50 |
huggingface/transformers | df8640cedb5dd3266ebf8958731f9f1315507182 | 203e27059b519bf12d6d503d7f4020a8a1089870 | [CLIPSeg] Make interpolate_pos_encoding default to True (#34419)
* Remove interpolate_pos_encoding
* Make fixup
* Make interpolate_pos_encoding default to True
* Reuse existing interpolation
* Add integration test | [
{
"path": "src/transformers/models/clipseg/modeling_clipseg.py",
"patch": "@@ -205,7 +205,7 @@ def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: int, width:\n \n return torch.cat((class_pos_embed, patch_pos_embed), dim=1)\n \n- def forward(self, pixel_values: torch.FloatTensor,... | 2024-10-31T21:15:04 |
nodejs/node | 0acdd8402614ce39a95a798b7e2898972332c4c7 | cb5f358ee7461f191db8c88e0af26023558f5232 | tools: fix test-npm-package
Migrate the script to the new common tmpDir API.
PR-URL: https://github.com/nodejs/node/pull/19293
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Myles Borins <myles.borins@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.... | [
{
"path": "tools/test-npm-package.js",
"patch": "@@ -22,6 +22,7 @@ const { createWriteStream, mkdirSync, rmdirSync } = require('fs');\n const path = require('path');\n \n const common = require('../test/common');\n+const tmpDir = require('../test/common/tmpdir');\n \n const projectDir = path.resolve(__dirna... | 2018-03-12T06:39:36 |
ollama/ollama | fdb109469f14631329a18bc782a381996842d0d6 | 05a43e078a89247dcc71c703c1bee2af97c1655d | llm: Allow overriding flash attention setting
As we automatically enable flash attention for more models, there
are likely some cases where we get it wrong. This allows setting
OLLAMA_FLASH_ATTENTION=0 to disable it, even for models that usually
have flash attention. | [
{
"path": "envconfig/config.go",
"patch": "@@ -145,8 +145,8 @@ func Remotes() []string {\n \treturn r\n }\n \n-func Bool(k string) func() bool {\n-\treturn func() bool {\n+func BoolWithDefault(k string) func(defaultValue bool) bool {\n+\treturn func(defaultValue bool) bool {\n \t\tif s := Var(k); s != \"\" ... | 2025-10-01T21:38:09 |
golang/go | ca2f15289337f57dcfb938000af249532f79e4d4 | dc122c7a9c45f1ae16125024d4f06953cc322bcd | [dev.typeparams] go/types: add missing test from dev.go2go
errors_test.go was missed during merging. Add it.
Change-Id: I321f08ae16ca02586875e1c7776f5d78f8690b4d
Reviewed-on: https://go-review.googlesource.com/c/go/+/289549
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
T... | [
{
"path": "src/go/types/errors_test.go",
"patch": "@@ -0,0 +1,25 @@\n+// Copyright 2020 The Go Authors. All rights reserved.\n+// Use of this source code is governed by a BSD-style\n+// license that can be found in the LICENSE file.\n+\n+package types\n+\n+import \"testing\"\n+\n+func TestStripAnnotations(t... | 2021-02-04T02:57:06 |
vercel/next.js | 264bce852461771e7d3e0adb85a493ed3fe2fe59 | ea1c8030d4c084a690e49c063c4dbc711da7000e | Do not treat dynamically imported chunks as evaluated (#426)
I didn't really understand this part of the code when I wrote this. The dynamic import is never an entrypoint, so we shouldn't treat it as evaluated. Note, the underlying chunk foo.js without the runtime code was always available. But, the client linked to t... | [
{
"path": "crates/turbopack-core/src/chunk/mod.rs",
"patch": "@@ -360,12 +360,12 @@ async fn chunk_content_internal<I: FromChunkableAsset>(\n false\n };\n if is_async {\n- if let Some((direct_chunk_item, fat_chunk... | 2022-09-27T19:40:05 |
huggingface/transformers | 203e27059b519bf12d6d503d7f4020a8a1089870 | c443d8d53685ae1d20b5f34e90c66312bd7b7a30 | Add image text to text pipeline (#34170)
* Standardize image-text-to-text-models-output
add post_process_image_text_to_text to chameleon and cleanup
Fix legacy kwarg behavior and deprecation warning
add post_process_image_text_to_text to qwen2_vl and llava_onevision
Add post_process_image_text_to_text to i... | [
{
"path": "docs/source/en/main_classes/pipelines.md",
"patch": "@@ -478,6 +478,12 @@ Pipelines available for multimodal tasks include the following.\n - __call__\n - all\n \n+### ImageTextToTextPipeline\n+\n+[[autodoc]] ImageTextToTextPipeline\n+ - __call__\n+ - all\n+\n ### MaskGenerationPipe... | 2024-10-31T19:48:11 |
ollama/ollama | 05a43e078a89247dcc71c703c1bee2af97c1655d | bc8909fb38525c89dda842d4ecfc86a933089a99 | fix panic on bootstrapDevices (#12475)
Wrong index variable was used. | [
{
"path": "discover/runner.go",
"patch": "@@ -456,7 +456,7 @@ func bootstrapDevices(ctx context.Context, ollamaLibDirs []string, extraEnvs []s\n \t\t\t\textra := strings.SplitN(extraEnvs[j], \"=\", 2)\n \t\t\t\tif cmp[0] == extra[0] {\n \t\t\t\t\tcmd.Env[i] = extraEnvs[j]\n-\t\t\t\t\textraDone[i] = true\n+\... | 2025-10-02T00:39:29 |
golang/go | 8869086d8f0a31033ccdc103106c768dc17216b1 | e491c6eea9ad599a0ae766a3217bd9a16ca3a25a | runtime: fix typo in histogram.go
indicies -> indices
Change-Id: Ia50ae5918fc7a53c23590a94a18087a99bfd9bb7
GitHub-Last-Rev: 98eb724275fd61d5f5ce5dad6b1010c10f76906d
GitHub-Pull-Request: golang/go#44095
Reviewed-on: https://go-review.googlesource.com/c/go/+/289529
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust:... | [
{
"path": "src/runtime/histogram.go",
"patch": "@@ -26,7 +26,7 @@ const (\n \t// The number of super-buckets (timeHistNumSuperBuckets), on the\n \t// other hand, defines the range. To reserve room for sub-buckets,\n \t// bit timeHistSubBucketBits is the first bit considered for\n-\t// super-buckets, so supe... | 2021-02-04T02:47:37 |
huggingface/transformers | c443d8d53685ae1d20b5f34e90c66312bd7b7a30 | 114dd812dd5192fe5b908abf2ea1b6cf4502a7c9 | Bug Fix for issue #34294 (#34295)
Update SiglipVisionEmbeddings.forward to cast input to correct dtype before embedding it. | [
{
"path": "src/transformers/models/siglip/modeling_siglip.py",
"patch": "@@ -308,7 +308,8 @@ def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: int, width:\n \n def forward(self, pixel_values: torch.FloatTensor, interpolate_pos_encoding=False) -> torch.Tensor:\n _, _, height, w... | 2024-10-31T17:51:15 |
vercel/next.js | f53f5815f2fd34b40fbb0a33c8daacdc37cb5d65 | b24800bebf2160bd0e655dc08a009daa2b019ec3 | Font loader support in app (#40898)
Makes sure font loader CSS ends up correctly in the Flight Manifest and Flight Client Entries.
## Bug
- [ ] Related issues linked using `fixes #number`
- [ ] Integration tests added
- [ ] Errors have a helpful link attached, see `contributing.md`
## Feature
- [ ] Implements an e... | [
{
"path": "packages/font/google/target.css",
"patch": "@@ -0,0 +1 @@\n+/* target file for webpack loader */",
"additions": 1,
"deletions": 0,
"language": "CSS"
},
{
"path": "packages/font/local/target.css",
"patch": "@@ -0,0 +1 @@\n+/* target file for webpack loader */",
"additio... | 2022-09-27T18:10:05 |
rust-lang/rust | ed75e2a79a752386099be07a21aabab0e8724945 | 8c7aa53a59bd9580e56da54540a4c01a16c20f95 | fix: Fix new nightly lints | [
{
"path": "src/tools/rust-analyzer/crates/hir-def/src/expr_store/lower.rs",
"patch": "@@ -1360,8 +1360,7 @@ impl ExprCollector<'_> {\n else {\n panic!(\"just expanded a macro, ExpansionSpanMap should be available\");\n };\n- let old_span_map... | 2025-03-31T11:26:52 |
golang/go | dc122c7a9c45f1ae16125024d4f06953cc322bcd | c910fd7b771cfbfc1b11a6eef750f835bf66c96c | [dev.typeparams] test: exclude a failing test again (fix 32bit builds)
Change-Id: I6290bc4921ef17586b5028d3f40a88372b175014
Reviewed-on: https://go-review.googlesource.com/c/go/+/289269
Trust: Robert Griesemer <gri@golang.org>
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Robert Griesemer <gri@golang.org>
Revie... | [
{
"path": "test/run.go",
"patch": "@@ -1932,16 +1932,17 @@ var excluded = map[string]bool{\n \t\"typecheck.go\": true, // invalid function is not causing errors when called\n \t\"writebarrier.go\": true, // correct diagnostics, but different lines (probably irgen's fault)\n \n-\t\"fixedbugs/bug176.go\": ... | 2021-02-03T20:09:25 |
huggingface/transformers | 294c170ff97fa34ba6a8e79fd3645706fdf8ffa9 | b5919e12f74190a5f50a47f07cb708ceb029244d | feat: add benchmarks pg indexes (#34536)
* feat: add benchmarks pg indexes
* refactor: remove debug `df -h` | [
{
"path": ".github/workflows/benchmark.yml",
"patch": "@@ -63,7 +63,6 @@ jobs:\n commit_id=$GITHUB_SHA\r\n fi\r\n commit_msg=$(git show -s --format=%s | cut -c1-70)\r\n- df -h\r\n python3 benchmark/llama.py \"${{ github.head_ref || github.ref_name }}\" \"$c... | 2024-10-31T16:41:06 |
vercel/next.js | b24800bebf2160bd0e655dc08a009daa2b019ec3 | 9657f99eedc7dece0ef134e4709fba8251c3d14a | [edge] serialize custom config to middleware-manifest (#40881)
This PR serializes `regions` into `middleware-manifest.json`,
allowing to extend Edge Functions and Middleware for deployment
providers.
## Feature
- [ ] Implements an existing feature request or RFC. Make sure the
feature request has been accepte... | [
{
"path": "packages/next/build/analysis/get-page-static-info.ts",
"patch": "@@ -18,6 +18,7 @@ import { RSC_MODULE_TYPES } from '../../shared/lib/constants'\n export interface MiddlewareConfig {\n matchers: MiddlewareMatcher[]\n unstable_allowDynamicGlobs: string[]\n+ regions: string[] | string\n }\n \n... | 2022-09-27T17:30:15 |
nodejs/node | 22b68042590de93109dbc2a4ddaf78caa24c2306 | 148d16ab5b297739b125716f6dea9499b947fe47 | lib: define printErr() in script string
This commit moves the printErr() function, used by the
tick profiler processer, into the code string passed to
vm.runInThisContext().
PR-URL: https://github.com/nodejs/node/pull/19285
Fixes: https://github.com/nodejs/node/issues/19260
Reviewed-By: Michaël Zasso <targos@protonma... | [
{
"path": "lib/internal/v8_prof_processor.js",
"patch": "@@ -22,11 +22,6 @@ scriptFiles.forEach(function(s) {\n script += process.binding('natives')[s] + '\\n';\n });\n \n-// eslint-disable-next-line no-unused-vars\n-function printErr(err) {\n- console.error(err);\n-}\n-\n const tickArguments = [];\n if ... | 2018-03-11T14:33:42 |
ollama/ollama | 35ac4eb12c35644ce8bca38dfc5d6636d2cafa16 | 3d0b1734c006798960a56acb0ea23ea57e0dd1d9 | fix keep alive
this reference to keep alive was missed in #12041 so chat has a
diffferent behaviour than generate | [
{
"path": "server/routes.go",
"patch": "@@ -1777,7 +1777,7 @@ func (s *Server) ChatHandler(c *gin.Context) {\n \t}\n \n \t// expire the runner\n-\tif len(req.Messages) == 0 && req.KeepAlive != nil && int(req.KeepAlive.Seconds()) == 0 {\n+\tif len(req.Messages) == 0 && req.KeepAlive != nil && req.KeepAlive.D... | 2025-10-01T00:12:37 |
golang/go | 5ef7357b50015e91b096a4f14f554db78ba18b5f | 671dba6c89bc0183982553e0359e0a3b2c00963f | [dev.fuzz] internal/fuzz: refactor CorpusEntry type
CorpusEntry is now a struct type with Name and Data fields. In the
future, it may have more fields describing multiple values with
different types added with f.Add.
CorpusEntry must be the same type in testing and
internal/fuzz. However, we don't want to export it f... | [
{
"path": "src/internal/fuzz/fuzz.go",
"patch": "@@ -39,7 +39,7 @@ import (\n //\n // If a crash occurs, the function will return an error containing information\n // about the crash, which can be reported to the user.\n-func CoordinateFuzzing(ctx context.Context, parallel int, seed [][]byte, corpusDir, cac... | 2021-02-01T23:00:37 |
huggingface/transformers | b5919e12f74190a5f50a47f07cb708ceb029244d | 4ca004eac66faec24591b59f594bbbd4c08c5acc | fix(DPT,Depth-Anything) Address expected_slice errors inside inference tests (#34518)
* fix(DPT,Depth-Anything) Address expected_slice errors inside inference tests
Signed-off-by: Phillip Kuznetsov <philkuz@gimletlabs.ai>
* [run_slow] dpt, depth_anything
---------
Signed-off-by: Phillip Kuznetsov <philkuz@... | [
{
"path": "tests/models/depth_anything/test_modeling_depth_anything.py",
"patch": "@@ -263,7 +263,7 @@ def test_inference(self):\n self.assertEqual(predicted_depth.shape, expected_shape)\n \n expected_slice = torch.tensor(\n- [[8.8204, 8.6468, 8.6195], [8.3313, 8.6027, 8.7526], [8... | 2024-10-31T15:47:58 |
vercel/next.js | c69677a553417930b2eaef9b76112ed468837c14 | 6be2868787285c878d4b4b9eeca2079c9f08bb7b | remove reducer from server bundle (#40959)
This reducer takes a good chunk of the bundle but is never run on the server so we can eliminate it. 5-10ms wins from my manual benchmark runs.
## Bug
- [ ] R... | [
{
"path": "packages/next/client/components/reducer.ts",
"patch": "@@ -615,7 +615,7 @@ type AppRouterState = {\n /**\n * Reducer that handles the app-router state updates.\n */\n-export function reducer(\n+function clientReducer(\n state: Readonly<AppRouterState>,\n action: Readonly<\n | ReloadActi... | 2022-09-27T16:36:17 |
nodejs/node | da226eaa868bb04b62a36502339b1d1678142e94 | c9794880e89dbbecbbd85e4ee0980ed9c7ac0971 | test: fix assertion argument order
Fix the assertion argument order so that it will report "actual" and
"expected" correctly when the test fails.
Ref: https://github.com/nodejs/node/issues/19263
PR-URL: https://github.com/nodejs/node/pull/19264
Refs: https://github.com/nodejs/node/issues/19263
Reviewed-By: Richard L... | [
{
"path": "test/sequential/test-inspector-stop-profile-after-done.js",
"patch": "@@ -25,7 +25,7 @@ async function runTests() {\n 'Waiting for the debugger to disconnect...');\n await session.send({ method: 'Profiler.stop' });\n session.disconnect();\n- assert.strictEqual(0, (await child.expect... | 2018-03-09T18:05:36 |
rust-lang/rust | 76034a648d63f88d00e3b24f369ba356c6e52282 | 7310925076c797d67d3df026c6b22f9a73380661 | do not run EnvVars::cleanup if there was an interpreter error | [
{
"path": "src/tools/miri/src/eval.rs",
"patch": "@@ -459,8 +459,13 @@ pub fn eval_entry<'tcx>(\n ecx.handle_ice();\n panic::resume_unwind(panic_payload)\n });\n- // `Ok` can never happen.\n+ // `Ok` can never happen; the interpreter loop always exits with an \"error\"\n+ // (bu... | 2025-03-31T09:51:06 |
golang/go | 3db6e18468d9e5c8f5fcfece26b5b666f86e9742 | bb53a5ad43732893da095e82334fb9f0ea912878 | [dev.typeparams] test: enable more errorcheck tests
These newly enabled (not anymore excluded) tests pass now
that we run in -G=3 mode when using the new types2 based
noder.
Change-Id: I5e7304c8020f394b79737d67c750bebbe02bd502
Reviewed-on: https://go-review.googlesource.com/c/go/+/289109
Trust: Robert Griesemer <gri@... | [
{
"path": "test/run.go",
"patch": "@@ -1932,18 +1932,16 @@ var excluded = map[string]bool{\n \t\"typecheck.go\": true, // invalid function is not causing errors when called\n \t\"writebarrier.go\": true, // correct diagnostics, but different lines (probably irgen's fault)\n \n-\t\"fixedbugs/bug176.go\": ... | 2021-02-03T05:50:40 |
huggingface/transformers | dca93ca076c68372dcf3ad1239a2119afdda629c | 1b86772de5d8429ff5ffe5caf7855548d2f4f072 | Fix step shifting when accumulate gradient (#33673)
* replace total_batched_samples with step while counting grad accum step
* remove unused variable
* simplify condition for update step
* fix format by ruff
* simplify update step condition using accelerator.sync_gradients
* simplify update condition us... | [
{
"path": "src/transformers/trainer.py",
"patch": "@@ -2404,7 +2404,6 @@ def _inner_training_loop(\n if args.eval_on_start:\n self._evaluate(trial, ignore_keys_for_eval, skip_scheduler=True)\n \n- total_batched_samples = 0\n for epoch in range(epochs_trained, num_train_epo... | 2024-10-31T13:53:23 |
ollama/ollama | 3d0b1734c006798960a56acb0ea23ea57e0dd1d9 | efaee8c2d658f7f40a2f44b411ebfb25fcc198b0 | ggml: Preallocate CUDA pool memory
The GGML CUDA backend allocates additional memory for intermediate
results during calculation. This memory isn't currently allocated
during worst case graph reservation and therefore not included in
scheduling. This means that as these buffers potentially grow
with context length, we... | [
{
"path": "llama/patches/0022-ggml-No-alloc-mode.patch",
"patch": "@@ -3,78 +3,102 @@ From: Jesse Gross <jesse@ollama.com>\n Date: Wed, 23 Jul 2025 11:58:49 -0700\n Subject: [PATCH] ggml: No-alloc mode\n \n-Callers can set a backend buffer type to be no-alloc, meaning that\n+Callers can set a scheduler to b... | 2025-09-09T23:17:31 |
vercel/next.js | 6be2868787285c878d4b4b9eeca2079c9f08bb7b | bb42d40fb74a1f2f277f29546ceba2003dae96b7 | misc: add benchmarking script for edge rendering (#40716)
This PR adds the benchmarking script I've been using for #40251 to
measure the performance improvements that we make to the Edge SSR
runtime.
This tool:
- uploads two version of the benchmarking project to Vercel, one with
the latest canary and the one w... | [
{
"path": ".gitignore",
"patch": "@@ -45,4 +45,4 @@ test-timings.json\n \n # Cache\n *.tsbuildinfo\n-.swc/\n+.swc/\n\\ No newline at end of file",
"additions": 1,
"deletions": 1,
"language": "Unknown"
},
{
"path": "bench/vercel/.env.dev",
"patch": "@@ -0,0 +1,8 @@\n+# The Vercel team... | 2022-09-27T15:57:16 |
huggingface/transformers | 1b86772de5d8429ff5ffe5caf7855548d2f4f072 | f38531619ddff23a510d5f7ccbc257a1bb1a3cb7 | Fix: img size mismatch caused by incorrect unpadding in LLaVA-Next (#34522)
Fix: unpadding img mismatch | [
{
"path": "src/transformers/models/llava_next/modeling_llava_next.py",
"patch": "@@ -138,12 +138,12 @@ def unpad_image(tensor, original_size):\n \n if original_aspect_ratio > current_aspect_ratio:\n scale_factor = current_width / original_width\n- new_height = int(original_height * scale_... | 2024-10-31T13:32:45 |
golang/go | e491c6eea9ad599a0ae766a3217bd9a16ca3a25a | fca94ab3ab113ceddb7934f76d0f1660cad98260 | math/big: fix comment in divRecursiveStep
There appears to be a typo in the description of
the recursive division algorithm.
Two things seem suspicious with the original comment:
1. It is talking about choosing s, but s doesn't
appear anywhere in the equation.
2. The math in the equation is incorrect.
Where... | [
{
"path": "src/math/big/nat.go",
"patch": "@@ -881,7 +881,7 @@ func (z nat) divRecursiveStep(u, v nat, depth int, tmp *nat, temps []*nat) {\n \t\t// then floor(u1/v1) >= floor(u/v)\n \t\t//\n \t\t// Moreover, the difference is at most 2 if len(v1) >= len(u/v)\n-\t\t// We choose s = B-1 since len(v)-B >= B+1... | 2021-01-27T15:33:35 |
nodejs/node | c9794880e89dbbecbbd85e4ee0980ed9c7ac0971 | 8403f00dc3e8f1c3c5cae4c4aec2eb10733cc15c | stream: make virtual methods errors consistent
Use the same error code and always emit the error instead of
throwing it.
PR-URL: https://github.com/nodejs/node/pull/18813
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Michaë Zasso <targos@protonmail.c... | [
{
"path": "lib/_http_outgoing.js",
"patch": "@@ -616,7 +616,7 @@ OutgoingMessage.prototype.removeHeader = function removeHeader(name) {\n \n \n OutgoingMessage.prototype._implicitHeader = function _implicitHeader() {\n- throw new ERR_METHOD_NOT_IMPLEMENTED('_implicitHeader()');\n+ this.emit('error', new E... | 2018-02-16T11:53:34 |
ollama/ollama | efaee8c2d658f7f40a2f44b411ebfb25fcc198b0 | 734b57da0e04d363e70ea47a3a6f372c0b9dac82 | ggml: Backport scale kernel fixes
The GGML scale kernel uses signed 32-bit ints to represent
the number of elements in the tensor. For large images,
mistral-small3.2 overflows this, triggering CUDA errors due
to negative arguments.
Currently, this can happen when the user passes a large image
to mistral-small3.2. How... | [
{
"path": "llama/patches/0026-ggml-Backport-scale-kernel-fixes.patch",
"patch": "@@ -0,0 +1,57 @@\n+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001\n+From: Jesse Gross <jesse@ollama.com>\n+Date: Tue, 23 Sep 2025 15:41:58 -0700\n+Subject: [PATCH] ggml: Backport scale kernel fixes\n+\n+... | 2025-09-23T19:13:39 |
vercel/next.js | bb42d40fb74a1f2f277f29546ceba2003dae96b7 | 833a67ff8cee3e14c84af11bf78b25085d643824 | Always show node_modules inside test in VS Code (#40956)
As per #40926 we commit all node_modules inside test so we can test external packages. It will be much more convenient to not hide them in VS Code. With this configuration, all node_modules folders will be excluded except for these under the test directory.
## ... | [
{
"path": ".vscode/settings.json",
"patch": "@@ -5,5 +5,10 @@\n { \"language\": \"typescript\", \"autoFix\": true },\n { \"language\": \"typescriptreact\", \"autoFix\": true }\n ],\n- \"debug.javascript.unmapMissingSources\": true\n+ \"debug.javascript.unmapMissingSources\": true,\n+ \"files.ex... | 2022-09-27T15:21:30 |
rust-lang/rust | 674a7adf9bd765354f58c50f3632f3410a6ac9f4 | 4dca28cfa27661e0d87f9de23761445cac62d2d6 | Add an error when full metadata was not found | [
{
"path": "compiler/rustc_metadata/messages.ftl",
"patch": "@@ -97,6 +97,10 @@ metadata_found_staticlib =\n found staticlib `{$crate_name}` instead of rlib or dylib{$add_info}\n .help = please recompile that crate using --crate-type lib\n \n+metadata_full_metadata_not_found =\n+ only metadata stu... | 2025-03-14T10:45:22 |
huggingface/transformers | 48872fd6ae336fbde6fac7706910a9a4bc48210e | 9f06fb05059a973048f5865e7e385c9db5d6daa4 | Add Image Processor Fast RT-DETR (#34354)
* add fast image processor rtdetr
* add gpu/cpu test and fix docstring
* remove prints
* add to doc
* nit docstring
* avoid iterating over images/annotations several times
* change torch typing
* Add image processor fast documentation | [
{
"path": "docs/source/en/main_classes/image_processor.md",
"patch": "@@ -18,6 +18,49 @@ rendered properly in your Markdown viewer.\n \n An image processor is in charge of preparing input features for vision models and post processing their outputs. This includes transformations such as resizing, normalizat... | 2024-10-30T17:49:47 |
nodejs/node | 8403f00dc3e8f1c3c5cae4c4aec2eb10733cc15c | 33e63fe64f238227481b83fc401a1a2371013995 | http2: fixes error handling
There should be no default error handling when using Http2Stream.
All errors will end up in `'streamError'` on the server anyway,
but they are emitted on `'stream'` as well, otherwise some error
conditions are impossible to debug.
See: https://github.com/nodejs/node/pull/14991
PR-URL: htt... | [
{
"path": "lib/internal/http2/core.js",
"patch": "@@ -2073,19 +2073,12 @@ function afterOpen(session, options, headers, streamOptions, err, fd) {\n headers, streamOptions));\n }\n \n-function streamOnError(err) {\n- // we swallow the error for parity with HTTP1\n- // all the e... | 2018-03-08T15:22:23 |
ollama/ollama | 734b57da0e04d363e70ea47a3a6f372c0b9dac82 | 0469861d9dfdec71fe895a0d4672056dfe8af54f | ggml: Remove allocation status reporting
For each memory allocation we report the size of the (attempted)
allocation and whether it succeeded or failed. The latter status
reporting proved to be not that useful in practice as systems
such as Windows can automatically overflow from VRAM into RAM,
resultings in successfu... | [
{
"path": "llama/patches/0014-graph-memory-reporting-on-failure.patch",
"patch": "@@ -4,48 +4,38 @@ Date: Fri, 18 Apr 2025 15:58:19 -0700\n Subject: [PATCH] graph memory reporting on failure\n \n ---\n- ggml/include/ggml-alloc.h | 6 ++++++\n- ggml/include/ggml-backend.h | 6 ++++++\n- ggml/src/ggml-alloc... | 2025-09-23T00:27:03 |
rust-lang/rust | a451b04d0140a25d8b55393b634be35e624021f4 | a2083a54e5f38448e4d0849a95620bd449ee1767 | fix: Cleanup param name inlay hint filtering | [
{
"path": "src/tools/rust-analyzer/crates/hir/src/attrs.rs",
"patch": "@@ -260,7 +260,7 @@ fn resolve_impl_trait_item(\n // attributes here. Use path resolution directly instead.\n //\n // FIXME: resolve type aliases (which are not yielded by iterate_path_candidates)\n- method_resolution::ite... | 2025-03-31T05:43:08 |
vercel/next.js | 833a67ff8cee3e14c84af11bf78b25085d643824 | b508fef2186c633bf9e01b01af141077f5c17b6c | Rename 404 -> not-found for new router (#40941)
As discussed with @sebmarkbage, the handling is more about showing a not found component than it is about a specific status code as these can come in late with streaming.
## Bug
- [ ] Related issues linked using `fixes #number`
- [ ] Integration tests added
- [ ] Err... | [
{
"path": "packages/next/build/webpack/loaders/next-app-loader.ts",
"patch": "@@ -8,7 +8,7 @@ export const FILE_TYPES = {\n template: 'template',\n error: 'error',\n loading: 'loading',\n- '404': '404',\n+ 'not-found': 'not-found',\n } as const\n \n // TODO-APP: check if this can be narrowed.\n@@ -9... | 2022-09-27T13:50:21 |
huggingface/transformers | 9f06fb05059a973048f5865e7e385c9db5d6daa4 | 5251fe6271bec670f71a6c1a86f4a2049fb03a90 | Fix super tiny extra space typo (#34440)
Update training_args.py | [
{
"path": "src/transformers/training_args.py",
"patch": "@@ -1030,7 +1030,7 @@ class TrainingArguments:\n use_cpu: bool = field(\n default=False,\n metadata={\n- \"help\": \" Whether or not to use cpu. If set to False, we will use cuda/tpu/mps/npu device if available.\"\n+ ... | 2024-10-30T15:55:16 |
nodejs/node | 0048169f5e6d280ffd4efd9ce3812bb09ff7309e | 90e70b8caf26b73241c6f002c53d8cc96b6496b5 | test: fix path in doctool/test-doctool-json
PR-URL: https://github.com/nodejs/node/pull/19287
Refs: https://github.com/nodejs/node/pull/17820
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com> | [
{
"path": "test/doctool/test-doctool-json.js",
"patch": "@@ -1,9 +1,9 @@\n 'use strict';\n \n const common = require('../common');\n-// The doctool currently uses js-yaml from the tool/eslint/ tree.\n+// The doctool currently uses js-yaml from the tool/node_modules/eslint/ tree.\n try {\n- require('../../t... | 2018-03-11T19:37:56 |
ollama/ollama | 83021fcf0fe935ee5c6d29b12cc357b373ec926f | 0469861d9dfdec71fe895a0d4672056dfe8af54f | qwen3-coder: fix tool definition type rendering | [
{
"path": "model/renderers/qwen3coder.go",
"patch": "@@ -99,9 +99,7 @@ func Qwen3CoderRenderer(messages []api.Message, tools []api.Tool, _ *api.ThinkVa\n \t\t\t\t\tsb.WriteString(\"\\n<name>\" + name + \"</name>\")\n \n \t\t\t\t\tif len(prop.Type) > 0 {\n-\t\t\t\t\t\t// TODO(!!!)(drifkin): we should match t... | 2025-09-30T22:03:15 |
huggingface/transformers | 5251fe6271bec670f71a6c1a86f4a2049fb03a90 | eab6c491d439e83d5e31c660df6f7e36592eb0a2 | Add GGUF for Mamba (#34200)
* add mamba architecture for gguf
* add logic for weights conversion, some fixes and refactoring
* add lm_head layers, unit test refactoring
* more fixes for tests
* remove lm_head creation
* remove unused comments | [
{
"path": "docs/source/en/gguf.md",
"patch": "@@ -86,6 +86,7 @@ For now the supported model architectures are the architectures that have been v\n - GPT2\n - Starcoder2\n - T5\n+- Mamba\n \n ## Example usage\n ",
"additions": 1,
"deletions": 0,
"language": "Markdown"
},
{
"path": "src/tr... | 2024-10-30T15:52:17 |
vercel/next.js | b508fef2186c633bf9e01b01af141077f5c17b6c | c86a9b992d0df0f29eb2a82f331261910677f838 | Merge e2e test node_modules (#40926)
Merge e2e tests customized `node_modules` with installed `node_modules`,
to let you debug easily locally without moving folder between
`node_modules` and `node_modules_bak`
Also add `optoutServerComponentsBundle` to config schema. | [
{
"path": "packages/next/server/config-schema.ts",
"patch": "@@ -240,6 +240,12 @@ const configSchema = {\n appDir: {\n type: 'boolean',\n },\n+ optoutServerComponentsBundle: {\n+ items: {\n+ type: 'string',\n+ },\n+ type: 'array',\n+ ... | 2022-09-27T13:18:08 |
nodejs/node | 98a14e026bab0bda4013f75d1d0968a12fe5117f | d3f174faab55662226ced53bef6bb2040352a825 | tls: expose Finished messages in TLSSocket
Exposes SSL_get_finished and SSL_get_peer_finished routines in OpenSSL
as tlsSocket.getFinished and tlsSocket.getPeerFinished, respectively.
PR-URL: https://github.com/nodejs/node/pull/19102
Fixes: https://github.com/nodejs/node/issues/19055
Refs: https://github.com/ripple/r... | [
{
"path": "doc/api/tls.md",
"patch": "@@ -583,6 +583,23 @@ if called on a server socket. The supported types are `'DH'` and `'ECDH'`. The\n \n For Example: `{ type: 'ECDH', name: 'prime256v1', size: 256 }`\n \n+### tlsSocket.getFinished()\n+<!-- YAML\n+added: REPLACEME\n+-->\n+\n+* Returns: {Buffer|undefine... | 2018-03-02T19:46:34 |
huggingface/transformers | 241d79026f1030124dbb957de936b3d617b621f2 | 8a734ea2c340beee23e665601919814918bf4c43 | fix pixtral processor (#34486)
* fix pixtral processor
* test out full length batches + remove undue ValueError
* fix up processing
* fix tests
* fix
* last fixup
* style
* [run-slow] pixtral
* [run-slow] pixtral
* fix config key
* skip torchscript tests
* [run-slow] pixtral
* add m... | [
{
"path": "src/transformers/models/pixtral/configuration_pixtral.py",
"patch": "@@ -52,6 +52,8 @@ class PixtralVisionConfig(PretrainedConfig):\n Dropout probability for the attention layers.\n rope_theta (`float`, *optional*, defaults to 10000.0):\n The base period of the RoP... | 2024-10-30T13:17:20 |
vercel/next.js | ea1c8030d4c084a690e49c063c4dbc711da7000e | 47e84c35a9d563698d0f8c0ad118f8da2e5dced0 | Fix a couple of typos (#417) | [
{
"path": "crates/turbopack-ecmascript/src/references/esm/export.rs",
"patch": "@@ -64,7 +64,7 @@ async fn expand_star_exports(root_asset: EcmascriptChunkPlaceableVc) -> Result<S\n EcmascriptExports::None => AnalyzeIssue {\n code: None,\n message: StringVc::cell(f... | 2022-09-27T12:37:40 |
nodejs/node | 695590e386845bc11bb42aa7f26327f2fba6d2a5 | a7e298a4a2720de6d463fb143b51f320136eab42 | test: fix compiler warnings in callback-scope
Currently there are two compiler warnings generated from the addons test
callback-scope:
../binding.cc:44:10:
warning: 'Resolve' is deprecated [-Wdeprecated-declarations]
local->Resolve(v8::Undefined(isolate));
^
../../../../deps/v8/include/v8.h:3893:45:
note: ... | [
{
"path": "test/addons/callback-scope/binding.cc",
"patch": "@@ -41,15 +41,17 @@ static void Callback(uv_work_t* req, int ignored) {\n \n v8::Local<v8::Promise::Resolver> local =\n v8::Local<v8::Promise::Resolver>::New(isolate, persistent);\n- local->Resolve(v8::Undefined(isolate));\n+ local->Reso... | 2018-03-09T07:29:27 |
huggingface/transformers | 8a734ea2c340beee23e665601919814918bf4c43 | 913330ca9f80b0a308d7490a02274b01b51e6051 | Tests: move `generate` tests to the right mixin and delete redundant tests (#34464)
* tmp commit
* tmp commit
* cull overwrites of deleted tests
* typo
* more specific docstring
* make fixup
* parameterize at the top?
* correction
* more deletions :D
* tmp commit
* for VLMs too
* fix _... | [
{
"path": "src/transformers/generation/utils.py",
"patch": "@@ -378,10 +378,14 @@ def prepare_inputs_for_generation(\n # If we have cache: let's slice `input_ids` through `cache_position`, to keep only the unprocessed tokens\n # Exception 1: when passing input_embeds, input_ids may be missin... | 2024-10-30T10:59:08 |
vercel/next.js | c86a9b992d0df0f29eb2a82f331261910677f838 | da8d299111ade846066035bc28c8c6ed38d1c400 | Code refactoring for webpack-config (#40942)
Clean up some utils, global variables and RegExp usages from the places
I touched recently.
## Bug
- [ ] Related issues linked using `fixes #number`
- [ ] Integration tests added
- [ ] Errors have a helpful link attached, see `contributing.md`
## Feature
- [ ... | [
{
"path": "packages/next/build/utils.ts",
"patch": "@@ -9,6 +9,7 @@ import path from 'path'\n import { promises as fs } from 'fs'\n import { isValidElementType } from 'next/dist/compiled/react-is'\n import stripAnsi from 'next/dist/compiled/strip-ansi'\n+import browserslist from 'next/dist/compiled/browsers... | 2022-09-27T10:03:57 |
golang/go | 98f8454a73b569d81d1c5e167d7b68f22e2e3fea | 1426a571b79bfcb3c0339e2fd96c893cd1549af6 | cmd/link: don't decode type symbol in shared library in deadcode
In the linker's deadcode pass we decode type symbols for
interface satisfaction analysis. When linking against Go shared
libraries, the type symbol may come from a shared library, so it
doesn't have data in the current module being linked, so we cannot
d... | [
{
"path": "misc/cgo/testshared/shared_test.go",
"patch": "@@ -1063,3 +1063,11 @@ func TestGCData(t *testing.T) {\n \tgoCmd(t, \"build\", \"-linkshared\", \"./gcdata/main\")\n \trunWithEnv(t, \"running gcdata/main\", []string{\"GODEBUG=clobberfree=1\"}, \"./main\")\n }\n+\n+// Test that we don't decode type ... | 2021-02-01T18:36:50 |
nodejs/node | a7e298a4a2720de6d463fb143b51f320136eab42 | 603afe25c870efb030aed84c2e634652def1a3ec | src: init emit_env_nonstring_warning_
Currently there is no member initialiser for emit_env_nonstring_warning_
in the Environment constructor leading to undefined behaviour.
For a debug build, this memory would be initialized:
(lldb) memory read -f x -s 4 -c 1 &emit_env_nonstring_warning_
0x7fff5fbfe254: 0x00000001
B... | [
{
"path": "src/env.cc",
"patch": "@@ -96,6 +96,7 @@ Environment::Environment(IsolateData* isolate_data,\n trace_sync_io_(false),\n abort_on_uncaught_exception_(false),\n emit_napi_warning_(true),\n+ emit_env_nonstring_warning_(true),\n makecallback_cntr_(0),\n should_abort... | 2018-03-11T10:22:18 |
ollama/ollama | c47154c08d59c93a653a2a798885a4a29bff71ee | b04e46da3ebca69a2b1216b3943d8a463e8b4a14 | fix: correct condition for AMDGPU_TARGETS filtering logic (#12412) | [
{
"path": "CMakeLists.txt",
"patch": "@@ -98,10 +98,12 @@ check_language(HIP)\n if(CMAKE_HIP_COMPILER)\n set(HIP_PLATFORM \"amd\")\n \n- find_package(hip REQUIRED)\n if(NOT AMDGPU_TARGETS)\n+ find_package(hip REQUIRED)\n list(FILTER AMDGPU_TARGETS INCLUDE REGEX \"^gfx(900|94[012]|1... | 2025-09-26T18:38:47 |
vercel/next.js | da8d299111ade846066035bc28c8c6ed38d1c400 | a92a3b6d3a88bab99adb40a225e9da7fa093bf55 | Fix SWC loader ignore for the server layer when Babel is used (#40939)
Specific logic to handle the file transpilation on the server layer is
implemented in SWC (#40603). When Babel is enabled, that SWC transform
is ignored at the moment. In this PR we add an additional SWC pass after
Babel to handle that.
## Bu... | [
{
"path": "packages/next/build/webpack-config.ts",
"patch": "@@ -628,52 +628,59 @@ export default async function getBaseWebpackConfig(\n loggedIgnoredCompilerOptions = true\n }\n \n- const getBabelOrSwcLoader = () => {\n- if (useSWCLoader && config?.experimental?.swcTraceProfiling) {\n+ const get... | 2022-09-27T09:31:25 |
golang/go | 1426a571b79bfcb3c0339e2fd96c893cd1549af6 | 32e789f4fb45b6296b9283ab80e126287eab4db5 | cmd/link: fix off-by-1 error in findShlibSection
We want to find a section that contains addr. sect.Addr+sect.Size
is the exclusive upper bound.
Change-Id: If2cd6bdd6e03174680e066189b0f4bf9e2ba6630
Reviewed-on: https://go-review.googlesource.com/c/go/+/288592
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cher... | [
{
"path": "src/cmd/link/internal/ld/decodesym.go",
"patch": "@@ -279,7 +279,7 @@ func findShlibSection(ctxt *Link, path string, addr uint64) *elf.Section {\n \tfor _, shlib := range ctxt.Shlibs {\n \t\tif shlib.Path == path {\n \t\t\tfor _, sect := range shlib.File.Sections[1:] { // skip the NULL section\n-... | 2021-02-01T21:46:49 |
huggingface/transformers | 913330ca9f80b0a308d7490a02274b01b51e6051 | 0f764a5af77de9a8f74521ec2237dc79feb40a71 | VLMs: fix number of image tokens (#34332)
* fix
* fix tests
* add tests
* style
* style
* fix qwen after rebase
* fix video llava | [
{
"path": "src/transformers/models/chameleon/modeling_chameleon.py",
"patch": "@@ -1288,7 +1288,7 @@ def forward(\n if pixel_values is not None:\n image_tokens = self.get_image_tokens(pixel_values)\n n_image_tokens_in_text = (input_ids == self.vocabulary_mapping.image_token_i... | 2024-10-30T09:21:37 |
nodejs/node | 603afe25c870efb030aed84c2e634652def1a3ec | 599337f43e0e0d66263e69d70edab26b61d3038a | doc: fix some recent nits in assert.md
* add missing bottom reference
* fix possible typos
* fix ABC-order in references and sections
PR-URL: https://github.com/nodejs/node/pull/19284
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Minwoo Jung <minwoo@nodesource.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.co... | [
{
"path": "doc/api/assert.md",
"patch": "@@ -326,7 +326,7 @@ rejected. See [`assert.rejects()`][] for more details.\n When `assert.doesNotReject()` is called, it will immediately call the `block`\n function, and awaits for completion.\n \n-Besides the async nature to await the completion behaves identical t... | 2018-03-11T12:05:04 |
ollama/ollama | b04e46da3ebca69a2b1216b3943d8a463e8b4a14 | 34efbbd3f02c024fc3256ae7d7799abb9cb98e8f | bugfix: restore the current runOptions if loading fails in the CLI (#12402)
There are two bugs when using `/load <model>` for a model that doesn't exist, namely:
1. it will not restore the current model settings if the current model is a thinking model; and
2. it will crash is the current model is a non-thinking m... | [
{
"path": "cmd/cmd.go",
"patch": "@@ -1118,6 +1118,51 @@ type runOptions struct {\n \tShowConnect bool\n }\n \n+func (r runOptions) Copy() runOptions {\n+\tvar messages []api.Message\n+\tif r.Messages != nil {\n+\t\tmessages = make([]api.Message, len(r.Messages))\n+\t\tcopy(messages, r.Messages)\n+\t}\n+\n... | 2025-09-26T01:30:45 |
vercel/next.js | a92a3b6d3a88bab99adb40a225e9da7fa093bf55 | 406d69d4d9f7d14b9bf497a134f0151914b13964 | Handle loading returning undefined (#40938)
Since React 18 returning `undefined` from components is allowed which renders to `undefined` in the RSC response so we need a separate prop to know if the loading boundary was provided.
Thanks to @finn-orsini who reported this problem.
## Bug
- [ ] Related issues linked... | [
{
"path": "packages/next/client/components/layout-router.client.tsx",
"patch": "@@ -278,11 +278,14 @@ export function InnerLayoutRouter({\n function LoadingBoundary({\n children,\n loading,\n+ hasLoading,\n }: {\n children: React.ReactNode\n loading?: React.ReactNode\n+ hasLoading: boolean\n }): J... | 2022-09-27T08:46:30 |
huggingface/transformers | 25a9fc584acb09afecb08b6cfd74e705058bf2ac | cd277618d4dbcafff108739e46584fd0a5c8f872 | Fix format mistake in string repr of tokenizer objects (#34493)
* fix repr string format for tokenizer objects
The repr of tokenizer tokens looks confusing and just stupid, like this: `Tokenizer(...), added_tokens_decoder={1: ..., 2: ...}`. The dict that is the value of the added_tokens_decoder attribute is outside... | [
{
"path": "src/transformers/tokenization_utils_base.py",
"patch": "@@ -1687,8 +1687,8 @@ def __repr__(self) -> str:\n f\"{self.__class__.__name__}(name_or_path='{self.name_or_path}',\"\n f\" vocab_size={self.vocab_size}, model_max_length={self.model_max_length}, is_fast={self.is_fast... | 2024-10-30T09:03:41 |
golang/go | 32e789f4fb45b6296b9283ab80e126287eab4db5 | 0b6cfea6342a7d95f74bc9e273039236ebd7e64f | test: fix incorrectly laid out instructions in issue11656.go
CL 279423 introduced a regression in this test as it incorrectly laid
out various instructions. In the case of arm, the second instruction
was overwriting the first. In the case of 386, amd64 and s390x, the
instructions were being appended to the end of the ... | [
{
"path": "test/fixedbugs/issue11656.go",
"patch": "@@ -59,10 +59,10 @@ func f(n int) {\n \till := make([]byte, 64)\n \tswitch runtime.GOARCH {\n \tcase \"386\", \"amd64\":\n-\t\till = append(ill, 0x89, 0x04, 0x25, 0x00, 0x00, 0x00, 0x00) // MOVL AX, 0\n+\t\till = append(ill[:0], 0x89, 0x04, 0x25, 0x00, 0x0... | 2021-02-01T03:02:18 |
nodejs/node | b325b5b435bcde9a2580e8ed05044b5cb352a5fd | e0bd2f31e584d8b188148084f97e80bc1573e555 | test: fix test-abort-backtrace in shared lib build
When using shared lib build, the binary path in the stack frames points
to shared lib. Change the checking criteria in the test case to match
that.
Refs: https://github.com/nodejs/node/issues/18535
Signed-off-by: Yihong Wang <yh.wang@ibm.com>
PR-URL: https://github... | [
{
"path": "test/abort/test-abort-backtrace.js",
"patch": "@@ -19,7 +19,8 @@ if (process.argv[2] === 'child') {\n }\n \n if (!common.isWindows) {\n- if (!frames.some((frame) => frame.includes(`[${process.execPath}]`))) {\n+ const { getBinaryPath } = require('../common/shared-lib-util');\n+ ... | 2018-03-07T21:34:10 |
huggingface/transformers | 9bee9ff5db6e68fb31065898d7e924d07c1eb9c1 | e4449bb790db5cb27a6cb36581991cf5e75209a9 | Un-deprecate timeout arg in pipelines (#34382)
* Un-deprecate timeout
* Put "timeout" on the allowed list
* make fixup | [
{
"path": "src/transformers/pipelines/depth_estimation.py",
"patch": "@@ -1,4 +1,3 @@\n-import warnings\n from typing import List, Union\n \n from ..utils import (\n@@ -72,6 +71,9 @@ def __call__(self, inputs: Union[str, List[str], \"Image.Image\", List[\"Image.Imag\n A dictionary of argumen... | 2024-10-29T18:45:14 |
vercel/next.js | 406d69d4d9f7d14b9bf497a134f0151914b13964 | 83c25b74d6dd0ee5abc44ea540301c310020e088 | Fix bundling and module resolution in the server layer (#40818)
We currently resolve the `react-server` condition correctly inside
`externals` for the server layer, however that will cause the resolved
path to be external (as it is called "externals").
So we need a way to hook into the module resolution process t... | [
{
"path": "packages/next/build/webpack-config.ts",
"patch": "@@ -1001,29 +1001,10 @@ export default async function getBaseWebpackConfig(\n return `commonjs next/dist/lib/import-next-warning`\n }\n \n- const resolveWithReactServerCondition =\n- layer === WEBPACK_LAYERS.server\n- ? ge... | 2022-09-27T08:18:06 |
golang/go | ca6999e27c395a30edb277dbda9c5b3c5854aace | aca22bddf231c862a1d6c9d8af8eed804c329d22 | [dev.regabi] test: add a test for inlining closures
Add a test case for issue 43818. We don't want to mark as inlinable a
function with a closure that has an operation (such as OSELRECV2) that
we don't currently support for exporting. This test case fails to
compile without the fix for #43818.
Updates #43818
Change-... | [
{
"path": "test/closure7.go",
"patch": "@@ -0,0 +1,28 @@\n+// run\n+\n+// Copyright 2020 The Go Authors. All rights reserved.\n+// Use of this source code is governed by a BSD-style\n+// license that can be found in the LICENSE file.\n+\n+package main\n+\n+func g(f func()) {\n+}\n+\n+// Must have exportable... | 2021-01-31T18:05:03 |
nodejs/node | f3257dd3ead29e620b2bc096d54b07b12ec5c5a3 | 2f743261814d428b935136b64b416a2879c29364 | benchmark: fix benchmark for url
Rename different parameters with the same names to make it possible
to run tests for url benchmarks.
PR-URL: https://github.com/nodejs/node/pull/19084
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Khaidi Chu <i@2333.moe>
Reviewed-By: Shingo Inoue <leko.noor@gmail.... | [
{
"path": "benchmark/url/legacy-vs-whatwg-url-searchparams-parse.js",
"patch": "@@ -2,10 +2,10 @@\n const common = require('../common.js');\n const { URLSearchParams } = require('url');\n const querystring = require('querystring');\n-const inputs = require('../fixtures/url-inputs.js').searchParams;\n+const ... | 2018-03-02T13:23:16 |
huggingface/transformers | e4449bb790db5cb27a6cb36581991cf5e75209a9 | f55595b177d66c3f8d9d43c8cccd52d263580bd1 | fix incorrect warning (#34416) | [
{
"path": "src/transformers/processing_utils.py",
"patch": "@@ -874,12 +874,13 @@ class MyProcessingKwargs(ProcessingKwargs, CommonKwargs, TextKwargs, ImagesKwarg\n else:\n # kwargs is a flat dictionary\n for key in kwargs:\n- if key not in ModelProcessorKwargs... | 2024-10-29T18:08:42 |
vercel/next.js | eb424042f83f1a1a04746734c3a75251d3ff7489 | 4f70f3af5da4f3a7c260bce55e1a164ee2e952f6 | Ensure skipClientCache is honored for router.push (#40932)
Fixes: https://github.com/vercel/next.js/issues/40927
## Bug
- [x] Related issues linked using `fixes #number`
- [x] Integration tests added
- [ ] Errors have a helpful link attached, see `contributing.md` | [
{
"path": "packages/next/shared/lib/router/router.ts",
"patch": "@@ -1550,6 +1550,7 @@ export default class Router implements BaseRouter {\n locale: nextState.locale,\n isPreview: nextState.isPreview,\n hasMiddleware: isMiddlewareMatch,\n+ unstable_skipClientCache: options.uns... | 2022-09-27T06:30:49 |
golang/go | 26e29aa15a189b26d3b2400a594d329368e78e79 | 6ac91e460c294bda5a50e628b7556bf20525fa44 | cmd/link: disable TestPIESize if CGO isn't enabled
With CGO disabled, the test throws the following error:
elf_test.go:291: # command-line-arguments
loadinternal: cannot find runtime/cgo
Change-Id: Iaeb183562ab637c714240b49e73078bdb791b35b
GitHub-Last-Rev: f8fe9afad5611411966413d17cb5874f7b0018a0
GitHub-Pull-Req... | [
{
"path": "src/cmd/link/elf_test.go",
"patch": "@@ -226,6 +226,12 @@ func main() {\n \n func TestPIESize(t *testing.T) {\n \ttestenv.MustHaveGoBuild(t)\n+\n+\t// We don't want to test -linkmode=external if cgo is not supported.\n+\t// On some systems -buildmode=pie implies -linkmode=external, so just\n+\t//... | 2021-01-26T16:29:05 |
ollama/ollama | 05ba4ca1f4b356df50ed6eede0e2bcdc76b31fb8 | fbd82ba5bb35c42a6b09f5bd50ff1aa0690b9626 | parsers: fix unicode handling for qwen3-coder
When trimming whitespace at the end of every chunk, we were iterating
backwards over the string byte-by-byte instead of rune-by-rune.
As an example of how this can cause corruption, suppose we have the
multi-byte character ✅ (`"\u2705"`), which is represented in utf-8 as
... | [
{
"path": "model/parsers/qwen3coder.go",
"patch": "@@ -11,6 +11,7 @@ import (\n \t\"strconv\"\n \t\"strings\"\n \t\"unicode\"\n+\t\"unicode/utf8\"\n \n \t\"github.com/ollama/ollama/api\"\n \t\"github.com/ollama/ollama/logutil\"\n@@ -204,12 +205,21 @@ func overlap(s, delim string) int {\n }\n \n func trailin... | 2025-09-25T22:47:46 |
Subsets and Splits
Assembly Language GitHub Issues
Retrieves a sample of assembly-language related commits with their details, but doesn't provide meaningful analysis or patterns beyond basic filtering.
Swift Compiler Issues Analysis
Retrieves all training data for the Swift programming language repository, providing basic filtering but offering limited analytical insight beyond identifying relevant code examples.