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 |
|---|---|---|---|---|---|
huggingface/transformers | 3b7675b2b844b02d4821b827871a21ad16dd446c | fa21ead73db473d88f8eca1ec244aba776fd9047 | fix FA2 when using quantization (#28203) | [
{
"path": "src/transformers/models/falcon/modeling_falcon.py",
"patch": "@@ -617,11 +617,11 @@ def forward(\n # cast them back in float16 just to be sure everything works as expected.\n input_dtype = query_layer.dtype\n if input_dtype == torch.float32:\n+ if torch.is_autoc... | 2023-12-26T03:06:41 |
nodejs/node | 8780e1d79b2162f95a50ec350b045868e93cac47 | ab2b331f5e3204f7b9b6362c7502910ecde6acae | test: allow inspector to reopen with same port
Test checks that if you open the inspector with '0' (pick a random free
port), close it, then reopen it, you get a different port. However this
isn't necessarily true.
PR-URL: https://github.com/nodejs/node/pull/14320
Fixes: https://github.com/nodejs/node/issues/14316
Re... | [
{
"path": "test/parallel/test-inspector-open.js",
"patch": "@@ -68,7 +68,6 @@ function tryToCloseWhenClosed(msg) {\n function reopenAfterClose(msg) {\n assert.strictEqual(msg.cmd, 'url');\n const port = url.parse(msg.url).port;\n- assert.notStrictEqual(port, firstPort);\n ping(port, (err) => {\n ... | 2017-07-17T07:01:03 |
ggml-org/llama.cpp | 459c0c2c1a400f960d7b8e8d94d31a8426f80986 | be79d9fdd95ab8955527c4aaa67b90e8b9516718 | server: fix SSE and OpenAI compatibility for error messages when streaming (#16109)
* server: fix SSE and OpenAI compatibility for error messages when streaming
* server: remove obsolete event parameter and use required data fieldname instead | [
{
"path": "tools/server/server.cpp",
"patch": "@@ -4679,17 +4679,17 @@ int main(int argc, char ** argv) {\n json res_json = result->to_json();\n if (res_json.is_array()) {\n for (const auto & res : res_json) {\n- if (... | 2025-09-20T05:56:30 |
vuejs/vue | aef5b4e47811cb842315bd27d6919650da45279b | 584e89da4ab17e1ebdae0ae10be77ef9d230c7a0 | fix(ssr): should render 0 as valid value for style property with unit | [
{
"path": "src/platforms/web/server/modules/style.js",
"patch": "@@ -23,7 +23,8 @@ export function genStyle (style: Object): string {\n function normalizeValue(key: string, value: any): string {\n if (\n typeof value === 'string' ||\n- (typeof value === 'number' && noUnitNumericStyleProps[key])\n+ ... | 2019-01-12T07:07:17 |
golang/go | bce1e25b717e07e05174117af6907999b3f51285 | b7a5e7ae920f46936d152b973d88682c0f1d6d0d | image/png: fix some 32-bit int overflows
Fixes #38435
Change-Id: Ib9ae3cf7f338b2860a5688e448a125f257fe624e
Reviewed-on: https://go-review.googlesource.com/c/go/+/230219
Reviewed-by: Andrew Ekstedt <andrew.ekstedt@gmail.com>
Reviewed-by: Rob Pike <r@golang.org> | [
{
"path": "src/image/png/reader.go",
"patch": "@@ -163,8 +163,9 @@ func (d *decoder) parseIHDR(length uint32) error {\n \tif w <= 0 || h <= 0 {\n \t\treturn FormatError(\"non-positive dimension\")\n \t}\n-\tnPixels := int64(w) * int64(h)\n-\tif nPixels != int64(int(nPixels)) {\n+\tnPixels64 := int64(w) * in... | 2020-04-27T02:23:26 |
huggingface/transformers | fa21ead73db473d88f8eca1ec244aba776fd9047 | 29e7a1e1834f331a4916853ecd58549ed78235d6 | [`Awq`] Enable the possibility to skip quantization for some target modules (#27950)
* v1
* add docstring
* add tests
* add awq 0.1.8
* oops
* fix test | [
{
"path": "docker/transformers-all-latest-gpu/Dockerfile",
"patch": "@@ -56,7 +56,7 @@ RUN python3 -m pip install --no-cache-dir auto-gptq --extra-index-url https://hu\n RUN python3 -m pip install --no-cache-dir einops\n \n # Add autoawq for quantization testing\n-RUN python3 -m pip install --no-cache-dir h... | 2023-12-25T10:06:56 |
ggml-org/llama.cpp | be79d9fdd95ab8955527c4aaa67b90e8b9516718 | f432d8d83e7407073634c5e4fd81a3d23a10827f | llama-bench: add --devices and --list-devices support (#16039)
* * llama-bench: add --devices support
- Support --devices same as llama-server
- Provide for benchmarking different device combinations
- Include --list-devices like llama-server for convenience
* fix: field display ordering restored
* fix: integrated t... | [
{
"path": "tools/llama-bench/README.md",
"patch": "@@ -30,8 +30,10 @@ options:\n --delay <0...N> (seconds) delay between each test (default: 0)\n -o, --output <csv|json|jsonl|md|sql> output format printed to stdout (default: md)\n -oe, --output-err <csv|json|jsonl|md|sql> output f... | 2025-09-19T22:15:21 |
nodejs/node | ab2b331f5e3204f7b9b6362c7502910ecde6acae | a6973a38117de2e1b7681647160e55c8f9ad4050 | test: make test-tls-alert-handling more strict
Use `common.mustCall()` and `common.mustNotCall()` to more rigorously
check that functions (especially no-op error handlers) are called the
expected number of times in test-tls-alert-handling.
PR-URL: https://github.com/nodejs/node/pull/14650
Reviewed-By: Colin Ihrig <cj... | [
{
"path": "test/parallel/test-tls-alert-handling.js",
"patch": "@@ -5,33 +5,43 @@ if (!common.hasCrypto)\n common.skip('missing crypto');\n \n if (!common.opensslCli)\n- common.skip('node compiled without OpenSSL CLI.');\n+ common.skip('node compiled without OpenSSL CLI');\n \n const net = require('net'... | 2017-08-06T17:48:14 |
vuejs/vue | 4f61d5b73b2e96b9570959dd2a540bc439d0f3bd | bc97ff956e373be14b67da4c57db207c1c889706 | chore: fix flow | [
{
"path": "src/compiler/codegen/index.js",
"patch": "@@ -27,7 +27,7 @@ export class CodegenState {\n this.dataGenFns = pluckModuleFunction(options.modules, 'genData')\n this.directives = extend(extend({}, baseDirectives), options.directives)\n const isReservedTag = options.isReservedTag || no\n-... | 2019-01-12T05:42:09 |
golang/go | 8bcf2834afdf6a1f7937390903a41518715ef6f5 | 09630172d45c09e150bd1329e37b74611d4af830 | net/http/httputil: make Switching Protocol requests (e.g. Websockets) cancelable
Ensures that a canceled client request for Switching Protocols
(e.g. h2c, Websockets) will cause the underlying connection to
be terminated.
Adds a goroutine in handleUpgradeResponse in order to select on
the incoming client request's co... | [
{
"path": "src/net/http/httputil/reverseproxy.go",
"patch": "@@ -526,7 +526,20 @@ func (p *ReverseProxy) handleUpgradeResponse(rw http.ResponseWriter, req *http.R\n \t\tp.getErrorHandler()(rw, req, fmt.Errorf(\"internal error: 101 switching protocols response with non-writable body\"))\n \t\treturn\n \t}\n-... | 2020-04-26T09:11:35 |
huggingface/transformers | 29e7a1e1834f331a4916853ecd58549ed78235d6 | 68fa1e855bef7d77e227686543787d8e2c4463fc | [`Llava`] Fix llava index errors (#28032)
* fix llava index errors
* forward contrib credits from original implementation and fix
* better fix
* final fixes and fix all tests
* fix
* fix nit
* fix tests
* add regression tests
---------
Co-authored-by: gullalc <gullalc@users.noreply.github.co... | [
{
"path": "src/transformers/models/llava/modeling_llava.py",
"patch": "@@ -431,8 +431,11 @@ def forward(\n if past_key_values is not None and pixel_values is not None and input_ids.shape[1] == 1:\n # Retrieve the first layer to inspect the logits and mask out the hidden s... | 2023-12-22T16:47:38 |
rust-lang/rust | 0d8015cb23a99a014e7bc82c19cd81e8ffa536a3 | da4c09937559e8d93cc6499ec8510a3571449061 | fix: handle character boundary in search mode | [
{
"path": "src/tools/rust-analyzer/crates/hir-def/src/import_map.rs",
"patch": "@@ -320,7 +320,7 @@ impl SearchMode {\n };\n match m {\n Some((index, _)) => {\n- name = &name[index + 1..];\n+ ... | 2025-01-12T20:31:15 |
vuejs/vue | 0d2e9c46f16e9ec5bd0f3eebd2aa31c7f7493856 | 743edacdb6fa0bb711e860b68373274f50c8baa5 | fix(core): dedupe lifecycle hooks during options merge
The fix landed in #9199 causes further extended constructors used as
mixins to drop options from its inheritance chain, so a different fix
is needed. | [
{
"path": "src/core/instance/init.js",
"patch": "@@ -117,32 +117,12 @@ export function resolveConstructorOptions (Ctor: Class<Component>) {\n function resolveModifiedOptions (Ctor: Class<Component>): ?Object {\n let modified\n const latest = Ctor.options\n- const extended = Ctor.extendOptions\n const... | 2019-01-11T22:51:45 |
ggml-org/llama.cpp | f432d8d83e7407073634c5e4fd81a3d23a10827f | 4067f07fc5aa5722b87db303b561597004696f6c | chat: Fix streaming parser for granite models (#15682)
* fix(chat): fix streaming parser for granite models
* tests: add test cases for Granite models chat parser | [
{
"path": "common/chat.cpp",
"patch": "@@ -2232,15 +2232,28 @@ static common_chat_params common_chat_params_init_granite(const common_chat_temp\n \n static void common_chat_parse_granite(common_chat_msg_parser & builder) {\n // Parse thinking tags\n+ static const common_regex start_think_regex(regex_... | 2025-09-19T15:57:30 |
nodejs/node | 73c59bbbf9bd2a53ab322cb822ae718fabef7b6c | 7307839b559b13c242e9d5341201ec17f83f3eed | deps: cherry-pick f19b889 from V8 upstream
Original commit message:
[inspector] support for cases when embedder doesn't call contextDestroyed
Node.js doesn't have good place to call contextDestroyed.
We need to cleanup everything on our side to allow clients to not call
contextDestroyed method.
R=dgozman@chromi... | [
{
"path": "deps/v8/src/inspector/inspected-context.cc",
"patch": "@@ -15,6 +15,39 @@\n \n namespace v8_inspector {\n \n+class InspectedContext::WeakCallbackData {\n+ public:\n+ WeakCallbackData(InspectedContext* context, V8InspectorImpl* inspector,\n+ int groupId, int contextId)\n+ :... | 2017-08-01T21:17:31 |
huggingface/transformers | 74d9d0cebb0263a3f8ab9c280569170cc74651d0 | 5da3db3fd5c070107df717a13382ccf1fe1efbe4 | Fixing visualization code for object detection to support both types of bounding box. (#27842)
* fix: minor enhancement and fix in bounding box visualization example
The example that was trying to visualize the bounding box was not considering an edge case,
where the bounding box can be un-normalized. So using th... | [
{
"path": "docs/source/en/tasks/object_detection.md",
"patch": "@@ -139,6 +139,17 @@ To get an even better understanding of the data, visualize an example in the dat\n ... box = annotations[\"bbox\"][i]\n ... class_idx = annotations[\"category\"][i]\n ... x, y, w, h = tuple(box)\n+... # Chec... | 2023-12-22T13:24:40 |
vuejs/vue | edf7df0c837557dd3ea8d7b42ad8d4b21858ade0 | 17d8bcb60edc14b2c23f5e6cc92f030897092e21 | fix(core): dedupe lifecycle hooks during options merge
The fix landed in #9199 causes further extended constructors used as
mixins to drop options from its inheritance chain, so a different fix
is needed. | [
{
"path": "src/core/instance/init.js",
"patch": "@@ -117,32 +117,12 @@ export function resolveConstructorOptions (Ctor: Class<Component>) {\n function resolveModifiedOptions (Ctor: Class<Component>): ?Object {\n let modified\n const latest = Ctor.options\n- const extended = Ctor.extendOptions\n const... | 2019-01-11T22:51:45 |
rust-lang/rust | a09ece29254cd06a1ee9529622e47c9a72645cb1 | da4c09937559e8d93cc6499ec8510a3571449061 | fix: Do not show safety hints for extern items lacking semantics | [
{
"path": "src/tools/rust-analyzer/crates/hir-def/src/dyn_map.rs",
"patch": "@@ -31,9 +31,9 @@ pub mod keys {\n \n use crate::{\n dyn_map::{DynMap, Policy},\n- BlockId, ConstId, EnumId, EnumVariantId, ExternCrateId, FieldId, FunctionId, ImplId,\n- LifetimeParamId, Macro2Id, MacroRu... | 2025-02-07T13:56:36 |
ggml-org/llama.cpp | 4067f07fc5aa5722b87db303b561597004696f6c | 4b8560ab56fdd9819358b47c338bbc8ec357c57e | feat: Improve mobile UI for Settings Dialog (#16084)
* feat: Improve mobile UI for Settings Dialog
* chore: update webui build output
* fix: Linting errors
* chore: update webui build output | [
{
"path": "tools/server/webui/scripts/install-git-hooks.sh",
"patch": "@@ -1,7 +1,7 @@\n #!/bin/bash\n \n # Script to install pre-commit and post-commit hooks for webui\n-# Pre-commit: formats code and builds, stashes unstaged changes\n+# Pre-commit: formats, lints, checks, and builds code, stashes unstaged... | 2025-09-19T07:52:27 |
nodejs/node | 7307839b559b13c242e9d5341201ec17f83f3eed | 2296b677fb1e2ea71e86a899b028ba9e817e86ad | test: check crypto before requiring tls module
test-tls-session-cache currently fails if built --without-ssl:
internal/util.js:82
throw new errors.Error('ERR_NO_CRYPTO');
^
Error [ERR_NO_CRYPTO]: Node.js is not compiled with OpenSSL crypto
support
at Object.assertCrypto (internal/util.js:82:11)
at tls... | [
{
"path": "test/parallel/test-tls-session-cache.js",
"patch": "@@ -21,6 +21,8 @@\n \n 'use strict';\n const common = require('../common');\n+if (!common.hasCrypto)\n+ common.skip('missing crypto');\n const fixtures = require('../common/fixtures');\n const assert = require('assert');\n const tls = require('... | 2017-08-09T05:28:28 |
golang/go | 09630172d45c09e150bd1329e37b74611d4af830 | 223d31f9fb1f821d0b1705fda3f925cff54d38b9 | net/http/httputil: fix typo in unit test name
Everywhere else is using "cancellation"
The reasoning is mentioned in 170060
> Though there is variation in the spelling of canceled,
> cancellation is always spelled with a double l.
>
> Reference: https://www.grammarly.com/blog/canceled-vs-cancelled/
Change-Id: Ifc97c... | [
{
"path": "src/net/http/httputil/reverseproxy_test.go",
"patch": "@@ -386,7 +386,7 @@ func TestReverseProxyFlushIntervalHeaders(t *testing.T) {\n \t}\n }\n \n-func TestReverseProxyCancelation(t *testing.T) {\n+func TestReverseProxyCancellation(t *testing.T) {\n \tconst backendResponse = \"I am the backend\"... | 2020-04-25T21:39:19 |
huggingface/transformers | 5da3db3fd5c070107df717a13382ccf1fe1efbe4 | c4df7c16685a340baa271bf139c1b3d60654f177 | [Whisper] Fix word-level timestamps with bs>1 or num_beams>1 (#28114)
* fix frames
* use smaller chunk length
* correct beam search + tentative stride
* fix whisper word timestamp in batch
* add test batch generation with return token timestamps
* Apply suggestions from code review
Co-authored-by: am... | [
{
"path": "src/transformers/models/whisper/modeling_whisper.py",
"patch": "@@ -2224,6 +2224,7 @@ def generate(\n if return_token_timestamps:\n kwargs[\"output_attentions\"] = True\n return_dict_in_generate = True\n+ kwargs[\"output_scores\"] = True\n \n ... | 2023-12-22T12:43:11 |
vuejs/vue | 743edacdb6fa0bb711e860b68373274f50c8baa5 | d21e93139697be2c0a6fdc4ee74d30d2834a729f | fix(core): fix merged twice bug when passing extended constructor to mixins (#9199)
fix #9198 | [
{
"path": "src/core/util/options.js",
"patch": "@@ -376,13 +376,13 @@ export function mergeOptions (\n }\n \n if (typeof child === 'function') {\n- child = child.options\n+ child = child.extendOptions\n }\n \n normalizeProps(child, vm)\n normalizeInject(child, vm)\n normalizeDirectives(chi... | 2019-01-10T21:10:54 |
ggml-org/llama.cpp | 4b8560ab56fdd9819358b47c338bbc8ec357c57e | 0dd58b6877f3dc106593d6bc68d98305f553c566 | chat : fix build on arm64 (#16101) | [
{
"path": "common/chat.cpp",
"patch": "@@ -1741,10 +1741,12 @@ static void common_chat_parse_gpt_oss(common_chat_msg_parser & builder) {\n static common_chat_params common_chat_params_init_firefunction_v2(const common_chat_template & tmpl, const struct templates_params & inputs) {\n LOG_DBG(\"%s\\n\", _... | 2025-09-19T06:02:51 |
huggingface/transformers | c4df7c16685a340baa271bf139c1b3d60654f177 | bb3bd4473929a63eefdc7c8cca9b1ebb634409d0 | Drop `feature_extractor_type` when loading an image processor file (#28195)
fix
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com> | [
{
"path": "src/transformers/image_processing_utils.py",
"patch": "@@ -75,6 +75,9 @@ class ImageProcessingMixin(PushToHubMixin):\n \n def __init__(self, **kwargs):\n \"\"\"Set elements of `kwargs` as attributes.\"\"\"\n+ # This key was saved while we still used `XXXFeatureExtractor` for im... | 2023-12-22T12:19:04 |
golang/go | 19648622d244f301f03d287b075e19d71d79864e | 7ef28cbd12d1777748affe9dccf20679665c06f9 | math/cmplx: fix typo in code comment
Everywhere else is using "cancellation" as of 2019
The reasoning is mentioned in 170060.
> Though there is variation in the spelling of canceled,
> cancellation is always spelled with a double l.
>
> Reference: https://www.grammarly.com/blog/canceled-vs-cancelled/
Change-Id: I93... | [
{
"path": "src/math/cmplx/sqrt.go",
"patch": "@@ -40,7 +40,7 @@ import \"math\"\n // 1/2\n // Im w = [ (r - x)/2 ] .\n //\n-// Cancelation error in r-x or r+x is avoided by using the\n+// Cancellation error in r-x or r+x is avoided by using the\n // identity 2 Re w Im w = y.\n /... | 2020-04-25T20:53:42 |
vuejs/vue | d21e93139697be2c0a6fdc4ee74d30d2834a729f | cf42f8e870f34e59e19f9cbf7ac6309c3c9c160c | fix(async component): memory leak after synchronous async loading (#9275)
fix #9229 | [
{
"path": "src/core/vdom/helpers/resolve-async-component.js",
"patch": "@@ -78,6 +78,8 @@ export function resolveAsyncComponent (\n // (async resolves are shimmed as synchronous during SSR)\n if (!sync) {\n forceRender(true)\n+ } else {\n+ contexts.length = 0\n }\n ... | 2019-01-11T20:46:25 |
nodejs/node | cea4bd92fd67bce63701fb708af708315ba88078 | c80d400c0128153322b35e69f813a8be5b5b3f14 | test: make test-tls-connect checks more strict
Check the error code on expected errors so that the introduction of
different errors in refactoring is caught.
While at it, re-order modules alphabetically per test-writing guide.
PR-URL: https://github.com/nodejs/node/pull/14695
Reviewed-By: Refael Ackermann <refack@gm... | [
{
"path": "test/parallel/test-tls-connect.js",
"patch": "@@ -25,9 +25,10 @@ const common = require('../common');\n if (!common.hasCrypto)\n common.skip('missing crypto');\n \n-const tls = require('tls');\n+const assert = require('assert');\n const fs = require('fs');\n const path = require('path');\n+cons... | 2017-08-08T16:53:55 |
ggml-org/llama.cpp | 69ffd891631befa9e6b485fd646a16dab4f2c007 | 246c0d9c795fb25da8268625d597580155a3672f | ggml-amx : fix ggml_amx_init() on generic Linux (#16049)
Generalize Linux check to `__linux__` to support non-glibc systems (like musl).
Also, return `false` on unknown/untested OS.
Without this commit, the code compiles (with warnings) but fails:
register_backend: registered backend CPU (1 devices)
register... | [
{
"path": "ggml/src/ggml-cpu/amx/amx.cpp",
"patch": "@@ -7,7 +7,7 @@\n #include \"ggml-cpu.h\"\n #include \"traits.h\"\n \n-#if defined(__gnu_linux__)\n+#if defined(__linux__)\n #include <sys/syscall.h>\n #include <unistd.h>\n #endif\n@@ -186,14 +186,16 @@ static size_t ggml_backend_amx_buffer_type_get_allo... | 2025-09-18T21:07:26 |
huggingface/transformers | bb3bd4473929a63eefdc7c8cca9b1ebb634409d0 | e37ab52dff8ba167938beac8f42349bda3198198 | Fix the check of models supporting FA/SDPA not run (#28202)
* add check_support_list.py
* fix
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com> | [
{
"path": ".circleci/config.yml",
"patch": "@@ -209,6 +209,7 @@ jobs:\n - run: python utils/update_metadata.py --check-only\n - run: python utils/check_task_guides.py\n - run: python utils/check_docstrings.py\n+ - run: python utils/check_support_list.py\n \n wo... | 2023-12-22T11:56:11 |
vuejs/vue | 17d8bcb60edc14b2c23f5e6cc92f030897092e21 | 7d9cfebe39ffd531e0948668e688474b276cdec1 | fix(ssr): should not render invalid numeric style values
Reverts part of 7d9cfebe - browsers only auto append units to
numbers when in quirksmode, so in standard mode, numbers set to
style properties that require units are invalid and should not
be rendered. | [
{
"path": "src/platforms/web/server/modules/style.js",
"patch": "@@ -21,18 +21,11 @@ export function genStyle (style: Object): string {\n }\n \n function normalizeValue(key: string, value: any): string {\n- if (typeof value === 'string') {\n+ if (\n+ typeof value === 'string' ||\n+ (typeof value ===... | 2019-01-11T19:54:10 |
golang/go | a9f8f02f3cfba18501669cdf58ae75ca36a4cff0 | 41e925bbcce2f0ca624f74ae4c5574dc76cb4140 | cmd/go/internal/cache: fix typing error in errVerifyMode
This change fixes the typing mistake in errVerifyMode error message in cache.
Change-Id: I10c405a06e3396f9932db72d9de418d7f8aa013c
GitHub-Last-Rev: 14ea7c693cdff599f0c6f64906fe46c2db62a87b
GitHub-Pull-Request: golang/go#38655
Reviewed-on: https://go-review.goog... | [
{
"path": "src/cmd/go/internal/cache/cache.go",
"patch": "@@ -108,7 +108,7 @@ const (\n // GODEBUG=gocacheverify=1.\n var verify = false\n \n-var errVerifyMode = errors.New(\"gocachverify=1\")\n+var errVerifyMode = errors.New(\"gocacheverify=1\")\n \n // DebugTest is set when GODEBUG=gocachetest=1 is in the... | 2020-04-25T05:42:10 |
nodejs/node | ef7da4200d29b5bfc815750eb80b515a0898f129 | 93f25f4b44e35b79f325a9bbddfa8d9d3210d02e | Revert "test: add DISABLED_ prefix to commented out test"
This reverts commit 75bf8a9db9e6005c6b5a5beb086a9def6e1c1b76.
Ref: https://github.com/nodejs/node/issues/14206
Ref: https://github.com/nodejs/node/pull/14317
PR-URL: https://github.com/nodejs/node/pull/14749
Reviewed-By: Refael Ackermann <refack@gmail.com>
Rev... | [
{
"path": "test/cctest/test_environment.cc",
"patch": "@@ -85,7 +85,8 @@ TEST_F(EnvironmentTest, AtExitWithArgument) {\n EXPECT_EQ(arg, cb_1_arg);\n }\n \n-TEST_F(EnvironmentTest, DISABLED_MultipleEnvironmentsPerIsolate) {\n+/*\n+TEST_F(EnvironmentTest, MultipleEnvironmentsPerIsolate) {\n const v8::Hand... | 2017-08-10T18:51:33 |
ggml-org/llama.cpp | 246c0d9c795fb25da8268625d597580155a3672f | 3edd87cd055a45d885fa914d879d36d33ecfc3e1 | cmake : fix static linking for OpenMP on Unix-like systems (#16031)
When compiling with GGML_STATIC=ON, the build process would produce a
binary that was still dynamically linked to OpenMP. This defeats the
purpose of a static build:
$ cmake -B build \
-DBUILD_SHARED_LIBS=OFF \
-DLLAMA_CUR... | [
{
"path": "ggml/src/CMakeLists.txt",
"patch": "@@ -114,6 +114,9 @@ message(STATUS \"GGML_SYSTEM_ARCH: ${GGML_SYSTEM_ARCH}\")\n \n if (NOT MSVC)\n if (GGML_STATIC)\n+ if (UNIX AND NOT APPLE)\n+ set(CMAKE_FIND_LIBRARY_SUFFIXES \".a;.so\")\n+ endif()\n add_link_options(-sta... | 2025-09-18T21:07:18 |
huggingface/transformers | e37ab52dff8ba167938beac8f42349bda3198198 | c9fb250a255cec542f9ec7bd072e3db2ebcbcd70 | Bug: `training_args.py` fix missing import with accelerate with version `accelerate==0.20.1` (#28171)
* fix-accelerate-version
* updated with exported ACCELERATE_MIN_VERSION,
* update string in ACCELERATE_MIN_VERSION | [
{
"path": "src/transformers/training_args.py",
"patch": "@@ -36,6 +36,7 @@\n SchedulerType,\n )\n from .utils import (\n+ ACCELERATE_MIN_VERSION,\n ExplicitEnum,\n cached_property,\n is_accelerate_available,\n@@ -1837,9 +1838,10 @@ def _setup_devices(self) -> \"torch.device\":\n r... | 2023-12-22T11:41:35 |
vuejs/vue | 7d9cfebe39ffd531e0948668e688474b276cdec1 | fe2b27ffb98932ec8e822bd88b4de1666810521c | fix(ssr): properly handle invalid and numeric style properties
Ignores values that are not string or numbers, and append px as default
unit to appropriate properties.
There will still be certain cases where the user simply provides an
invalid string value to a property which will be too costly to detect
(it's possibl... | [
{
"path": "src/platforms/web/server/modules/style.js",
"patch": "@@ -1,6 +1,6 @@\n /* @flow */\n \n-import { escape } from '../util'\n+import { escape, noUnitNumericStyleProps } from '../util'\n import { hyphenate } from 'shared/util'\n import { getStyle } from 'web/util/style'\n \n@@ -11,15 +11,34 @@ expor... | 2019-01-11T16:07:40 |
golang/go | b2fde1098a151f89c7a20d0fe322dfb052211786 | c7a11099c977b6e09c82ceaed014a4a55da568eb | [dev.link] cmd/link: use new dodata on darwin/amd64
This probably breaks darwin/arm64. Will fix.
Change-Id: I8be168985124f971e9d8ab5bc95c303336dd705b
Reviewed-on: https://go-review.googlesource.com/c/go/+/230019
Reviewed-by: Than McIntosh <thanm@google.com> | [
{
"path": "src/cmd/link/internal/ld/data.go",
"patch": "@@ -1355,8 +1355,7 @@ func (ctxt *Link) dodata2(symGroupType []sym.SymKind) {\n \t//\n \t// On darwin, we need the symbol table numbers for dynreloc.\n \tif ctxt.HeadType == objabi.Hdarwin {\n-\t\tpanic(\"not yet implemented for darwin\")\n-\t\t//\tmac... | 2020-04-24T18:57:20 |
nodejs/node | 93f25f4b44e35b79f325a9bbddfa8d9d3210d02e | b559e6625294be480f590a3502c57fff1f1106f9 | test: properly order freeing resources in cctest
The `IsolateData` instance is created before the `Environment` instance,
so free in reverse order.
Fixes: https://github.com/nodejs/node/issues/14206
PR-URL: https://github.com/nodejs/node/pull/14749
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Ben Noo... | [
{
"path": "test/cctest/test_environment.cc",
"patch": "@@ -41,9 +41,9 @@ class EnvironmentTest : public NodeTestFixture {\n }\n \n ~Env() {\n- FreeIsolateData(isolate_data_);\n environment_->CleanupHandles();\n FreeEnvironment(environment_);\n+ FreeIsolateData(isolate_data_);\n... | 2017-08-10T18:50:27 |
ggml-org/llama.cpp | 368560a1e3b9a3bc83af741b0b2bc9e46fb420d2 | 4ca088b036313c2d8e682f4cfeb7c29edd85d0b9 | CUDA: fix compilation on CC 6.0 (#16091) | [
{
"path": "ggml/src/ggml-cuda/fattn-tile.cu",
"patch": "@@ -35,7 +35,6 @@ static int fattn_tile_get_kq_stride_host(const int D, const int ncols, const int\n switch (D) {\n case 64:\n case 128:\n- return 128;\n case 256:\n return ncol... | 2025-09-18T17:28:32 |
huggingface/transformers | 1ef86c4f56e204384369059f574dc4a823fb7f07 | 548a8f6119e93f11469d54b6c1c1e7b9b73ec56d | Fix: [SeamlessM4T - S2TT] Bug in batch loading of audio in torch.Tensor format in the SeamlessM4TFeatureExtractor class (#27914)
* fixes: code fixes on is_batched condition to also check for batched audio data in torch.Tensor format instead of only just checking for batched audio data in np.ndarray format
* Update ... | [
{
"path": "src/transformers/models/seamless_m4t/feature_extraction_seamless_m4t.py",
"patch": "@@ -20,6 +20,12 @@\n \n import numpy as np\n \n+from ...utils import is_torch_available\n+\n+\n+if is_torch_available():\n+ import torch\n+\n from ...audio_utils import mel_filter_bank, spectrogram, window_func... | 2023-12-22T10:47:30 |
vuejs/vue | 537161779ea329c1d0a993997555f1c692b8cac1 | c50bbde41c4a1868a8a0b33df3238346840bd37c | fix(core): fix merged twice bug when passing extended constructor to mixins (#9199)
fix #9198 | [
{
"path": "src/core/util/options.js",
"patch": "@@ -382,13 +382,13 @@ export function mergeOptions (\n }\n \n if (typeof child === 'function') {\n- child = child.options\n+ child = child.extendOptions\n }\n \n normalizeProps(child, vm)\n normalizeInject(child, vm)\n normalizeDirectives(chi... | 2019-01-10T21:10:54 |
golang/go | c7a11099c977b6e09c82ceaed014a4a55da568eb | 1d083eba5bf09427ac132b913091cb3d8c1e733a | [dev.link] cmd/link: fix buglet in new GCProg generation code
It should check the name of the symbol being added, not the
GC data symbol we're generating.
Change-Id: I123679778ee542b8d1f5c15bf090fa3578025c19
Reviewed-on: https://go-review.googlesource.com/c/go/+/230018
Run-TryBot: Cherry Zhang <cherryyz@google.com>
R... | [
{
"path": "src/cmd/link/internal/ld/data.go",
"patch": "@@ -1084,14 +1084,14 @@ func (p *GCProg2) AddSym(s loader.Sym) {\n \t// Things without pointers should be in sym.SNOPTRDATA or sym.SNOPTRBSS;\n \t// everything we see should have pointers and should therefore have a type.\n \tif typ == 0 {\n-\t\tswitch... | 2020-04-24T19:19:55 |
nodejs/node | b559e6625294be480f590a3502c57fff1f1106f9 | 97c43940c81a835c21650a3323c5005a5b620e7c | src: add overlooked handle to cleanup
The `Environment::destroy_ids_timer_handle` should be cleaned up
by `Environment::CleanupHandles()`. Fix that by adding it to the list.
This partially fixes a cctest.
Ref: https://github.com/nodejs/node/issues/14206
PR-URL: https://github.com/nodejs/node/pull/14749
Reviewed-By: ... | [
{
"path": "src/env.cc",
"patch": "@@ -76,6 +76,10 @@ void Environment::Start(int argc,\n reinterpret_cast<uv_handle_t*>(&idle_check_handle_),\n close_and_finish,\n nullptr);\n+ RegisterHandleCleanup(\n+ reinterpret_cast<uv_handle_t*>(&destroy_ids_timer_handle_),\n+ close_and_fin... | 2017-08-10T18:48:18 |
ggml-org/llama.cpp | d304f459d8619399eb232b1e3689379306f439d3 | 0320ac5264279d74f8ee91bafa6c90e9ab9bbb91 | GGML WebGPU: Support for ADD, MUL, RMS_NORM, GET_ROWS operators (#16018)
* Add paramater buffer pool, batching of submissions, refactor command building/submission
* Add header for linux builds
* Free staged parameter buffers at once
* Format with clang-format
* Fix thread-safe implementation
* Use device implici... | [
{
"path": "ggml/src/ggml-webgpu/ggml-webgpu.cpp",
"patch": "@@ -116,6 +116,10 @@ struct webgpu_context_struct {\n wgpu::Queue queue;\n wgpu::Limits limits;\n \n+ // Separate this out from limits since on some Metal systems, the limit returned by\n+ // querying the limits is higher than th... | 2025-09-17T20:09:40 |
vuejs/vue | b32c4b693535c35ae10742eb3351cbc123b15941 | d855027741d4f602cc9f507c18ede75c3e7ce84a | fix: ensure proxied normal slot uses correct key | [
{
"path": "src/core/vdom/helpers/normalize-scoped-slots.js",
"patch": "@@ -19,7 +19,7 @@ export function normalizeScopedSlots (\n }\n // expose normal slots on scopedSlots\n for (const key in normalSlots) {\n- res[key] = () => normalSlots[key]\n+ res[key] = proxyNormalSlot(normalSlots, key)\n ... | 2019-01-10T17:32:06 |
huggingface/transformers | 548a8f6119e93f11469d54b6c1c1e7b9b73ec56d | 71f460578d5c85739caf42dc6f52e1c2eb12492d | Fix ONNX export for causal LM sequence classifiers by removing reverse indexing (#28144)
* normalize reverse indexing for causal lm sequence classifiers
* normalize reverse indexing for causal lm sequence classifiers
* normalize reverse indexing for causal lm sequence classifiers
* use modulo instead
* uni... | [
{
"path": "src/transformers/models/bloom/modeling_bloom.py",
"patch": "@@ -1011,7 +1011,10 @@ def forward(\n sequence_lengths = -1\n else:\n if input_ids is not None:\n- sequence_lengths = (torch.ne(input_ids, self.config.pad_token_id).sum(-1) - 1).to(logits.de... | 2023-12-22T10:33:44 |
nodejs/node | c6126b1308d5f7a8a06a36a3c1b523268edfd3fe | 0eb72684518b74c0b0d7d773cb71986b7418cb9a | test: refactor test-fs-stat
* add `use strict'
* change checks that `this` is mapped to `global` in sloppy mode to
checks that `this` is `undefined`
* modify arguments to assertions to match docs (actual first, expected
second)
* add blank line below `common` declaration per test writing guide
* use `assert.ifErro... | [
{
"path": "test/parallel/test-fs-stat.js",
"patch": "@@ -19,15 +19,19 @@\n // OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE\n // USE OR OTHER DEALINGS IN THE SOFTWARE.\n \n-/* eslint-disable strict */\n+'use strict';\n const common = require('../common');\n+\n const assert = requ... | 2017-08-06T00:25:30 |
vuejs/vue | 8b382b3efb14986e9ea5213d0b57c244b70ae0e7 | 5fb23d4e2971480d14fbee0146e3416a07bc0b9f | fix: allow _ in watch paths (element compat) | [
{
"path": "src/core/util/lang.js",
"patch": "@@ -30,7 +30,7 @@ export function def (obj: Object, key: string, val: any, enumerable?: boolean) {\n /**\n * Parse simple path.\n */\n-const bailRE = new RegExp(`[^${unicodeLetters}.$]`)\n+const bailRE = new RegExp(`[^${unicodeLetters}.$_]`)\n export function p... | 2019-01-10T04:46:38 |
golang/go | d422f54619b5b6e6301eaa3e9f22cfa7b65063c8 | 396833caef83b20f38199f9d74cb3e768b2fd478 | os, net: define and use os.ErrDeadlineExceeded
If an I/O operation fails because a deadline was exceeded,
return os.ErrDeadlineExceeded. We used to return poll.ErrTimeout,
an internal error, and told users to check the Timeout method.
However, there are other errors with a Timeout method that returns true,
notably sys... | [
{
"path": "doc/go1.15.html",
"patch": "@@ -172,6 +172,25 @@ <h3 id=\"minor_library_changes\">Minor changes to the library</h3>\n </dd>\n </dl>\n \n+<dl id=\"net\"><dt><a href=\"/pkg/net/\">net</a></dt>\n+ <dd>\n+ <p><!-- CL -->\n+ If an I/O operation exceeds a deadline set by\n+ the <a href=... | 2020-04-17T22:42:12 |
huggingface/transformers | 71f460578d5c85739caf42dc6f52e1c2eb12492d | 3a8769f6a9672bfe6e499923a8ebb6fb25d8393b | Update `docs/source/en/perf_infer_gpu_one.md` (#28198)
fix
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com> | [
{
"path": "docs/source/en/perf_infer_gpu_one.md",
"patch": "@@ -172,6 +172,8 @@ For now, Transformers supports SDPA inference and training for the following arc\n * [Llama](https://huggingface.co/docs/transformers/model_doc/llama#transformers.LlamaModel)\n * [Idefics](https://huggingface.co/docs/transformer... | 2023-12-22T09:40:22 |
rust-lang/rust | 80c60fe78375cb10a2fed6540dc2b29751fc4263 | c182ce9cbc8c29ebc1b4559d027df545e6cdd287 | std: replace the `FromInner` implementation for addresses with private conversion functions
Having these implementation available crate-wide means that platforms not using sockets for their networking code have to stub out the libc definitions required to support them. This PR moves the conversions to private helper f... | [
{
"path": "library/std/src/net/ip_addr.rs",
"patch": "@@ -8,32 +8,3 @@ pub use core::net::IpAddr;\n pub use core::net::Ipv6MulticastScope;\n #[stable(feature = \"rust1\", since = \"1.0.0\")]\n pub use core::net::{Ipv4Addr, Ipv6Addr};\n-\n-use crate::sys::net::netc as c;\n-use crate::sys_common::{FromInner, ... | 2025-02-07T17:34:09 |
ggml-org/llama.cpp | 0320ac5264279d74f8ee91bafa6c90e9ab9bbb91 | a7a98e0fffed794396b3fbad4dcdbbc184963645 | metal : refactor + optimize v2 (#15995)
* metal : improve naming
* metal : refactor device
ggml-ci
* cont : props
ggml-ci
* metal : apply ggml_mem_ranges_t
ggml-ci
* metal : remove GGML_METAL_USE_BF16
ggml-ci
* metal : refactor device buffer
ggml-ci
* cont : fix naming
* metal : sync before destroying the ... | [
{
"path": "ci/run.sh",
"patch": "@@ -45,7 +45,7 @@ SRC=`pwd`\n CMAKE_EXTRA=\"-DLLAMA_FATAL_WARNINGS=ON -DLLAMA_CURL=ON\"\n \n if [ ! -z ${GG_BUILD_METAL} ]; then\n- CMAKE_EXTRA=\"${CMAKE_EXTRA} -DGGML_METAL=ON -DGGML_METAL_USE_BF16=ON\"\n+ CMAKE_EXTRA=\"${CMAKE_EXTRA} -DGGML_METAL=ON\"\n fi\n \n if [ ... | 2025-09-17T17:38:12 |
vuejs/vue | 5fb23d4e2971480d14fbee0146e3416a07bc0b9f | 4c2c6f03dc617e0b7f8ffee7539876d131506cbf | fix: ensure only nromalize a scoped slot when it is present | [
{
"path": "src/core/vdom/helpers/normalize-scoped-slots.js",
"patch": "@@ -12,7 +12,9 @@ export function normalizeScopedSlots (\n } else {\n res = {}\n for (const key in slots) {\n- res[key] = normalizeScopedSlot(slots[key])\n+ if (slots[key]) {\n+ res[key] = normalizeScopedSlot(s... | 2019-01-10T03:47:11 |
huggingface/transformers | cd1350ce9bec0009ff37b3287a5acde064be93a7 | 260b9d2179ea9592b24ff102ab9ea672f6a4f3ef | Fix slow backbone tests - out_indices must match stage name ordering (#28186)
Indices must match stage name ordering | [
{
"path": "tests/models/auto/test_modeling_auto.py",
"patch": "@@ -244,8 +244,8 @@ def test_auto_backbone_timm_model_from_pretrained(self):\n self.assertIsInstance(model, TimmBackbone)\n \n # Check kwargs are correctly passed to the backbone\n- model = AutoBackbone.from_pretrained(\"r... | 2023-12-21T18:16:50 |
nodejs/node | 0eb72684518b74c0b0d7d773cb71986b7418cb9a | 52a0d97187f92579676dce54e05724606374c2fc | doc: fix http2 sample code for http2.md
PR-URL: https://github.com/nodejs/node/pull/14667
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: Luigi Pinca <luigipinca@gmai... | [
{
"path": "doc/api/http2.md",
"patch": "@@ -1242,7 +1242,7 @@ const {\n HTTP2_HEADER_CONTENT_TYPE\n } = http2.constants;\n \n-const server = http.createServer();\n+const server = http2.createServer();\n server.on('stream', (stream, headers, flags) => {\n const method = headers[HTTP2_HEADER_METHOD];\n ... | 2017-08-07T16:15:44 |
golang/go | 4a7e363288a055f97d2db4d96d7776e506aebcd5 | b3e8a00060030a8c60db3dbc0d2cf389c65c8a56 | cmd/compile: optimize Move with all-zero ro sym src to Zero
We set up static symbols during walk that
we later make copies of to initialize local variables.
It is difficult to ascertain at that time exactly
when copying a symbol is profitable vs locally
initializing an autotmp.
During SSA, we are much better placed t... | [
{
"path": "src/cmd/compile/internal/ssa/gen/generic.rules",
"patch": "@@ -2020,6 +2020,7 @@\n \t=> (Zero {t} [n] dst1 mem)\n (Move {t} [n] dst1 src mem:(VarDef (Zero {t} [n] dst0 _))) && isSamePtr(src, dst0)\n \t=> (Zero {t} [n] dst1 mem)\n+(Move {t} [n] dst (Addr {sym} (SB)) mem) && symIsROZero(sym) => (Ze... | 2020-04-23T20:28:14 |
vuejs/vue | 4c2c6f03dc617e0b7f8ffee7539876d131506cbf | 8eea6b07c415ac5c7e3a401265e6390f0e18e5b9 | build: fix dev:cjs script | [
{
"path": "package.json",
"patch": "@@ -15,7 +15,7 @@\n \"sideEffects\": false,\n \"scripts\": {\n \"dev\": \"rollup -w -c scripts/config.js --environment TARGET:web-full-dev\",\n- \"dev:cjs\": \"rollup -w -c scripts/config.js --environment TARGET:web-runtime-cjs\",\n+ \"dev:cjs\": \"rollup -w... | 2019-01-10T03:26:54 |
ggml-org/llama.cpp | 8f8f2274ee3601fecf6e2d57b52f701c81bede21 | c959b676be29e93f8dbc3bd6056ceba812a9eb72 | convert : add Llama4ForCausalLM (#16042)
* convert : add Llama4ForCausalLM
* handle swa
* half working version
* fix use_kq_norm
* fix use_kq_norm | [
{
"path": "convert_hf_to_gguf.py",
"patch": "@@ -2393,7 +2393,10 @@ def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iter\n return [] # skip other tensors\n \n \n-@ModelBase.register(\"Llama4ForConditionalGeneration\")\n+@ModelBase.register(\n+ \"Llama4ForConditionalGen... | 2025-09-17T17:18:21 |
huggingface/transformers | 260b9d2179ea9592b24ff102ab9ea672f6a4f3ef | f9a98c476c3a92beaba3b20e51f1ff49417231a6 | Even more TF test fixes (#28146)
* Fix vision text dual encoder
* Small cleanup for wav2vec2 (not fixed yet)
* Small fix for vision_encoder_decoder
* Fix SAM builds
* Update TFBertTokenizer test with modern exporting + tokenizer
* Fix DeBERTa
* Fix DeBERTav2
* Try RAG fix but it's impossible to te... | [
{
"path": "src/transformers/modeling_tf_pytorch_utils.py",
"patch": "@@ -57,7 +57,7 @@ def convert_tf_weight_name_to_pt_weight_name(\n transposed with regards to each other\n \"\"\"\n if name_scope is not None:\n- if not tf_name.startswith(name_scope):\n+ if not tf_name.start... | 2023-12-21T15:14:46 |
nodejs/node | 8172c8a5068b9e28c9a38d058417dd2254ab8eb6 | 4796fbc7d820240dde544b89755285e3af5bcf6c | test: set module loading error for aix
In test/parallel/test-module-loading-error.js, an attempt is made to
load a text file as a native executable. This results in an error
message in a platform specific manner.
AIX was not included in the list of platforms. This fix introduces
the AIX error messages.
PR-URL: https... | [
{
"path": "test/parallel/test-module-loading-error.js",
"patch": "@@ -28,7 +28,9 @@ const errorMessagesByPlatform = {\n win32: ['%1 is not a valid Win32 application'],\n linux: ['file too short', 'Exec format error'],\n sunos: ['unknown file type', 'not an ELF file'],\n- darwin: ['file too short']\n+... | 2017-07-27T13:40:15 |
golang/go | 1d083eba5bf09427ac132b913091cb3d8c1e733a | 42cca1a7fee916bb4ba769cb2db259ef6ec1c179 | [dev.link] cmd/link: fix minor error on error reporting
Correctly propagate ... arguments. (Maybe vet should warn on it?)
Reviewed-on: https://go-review.googlesource.com/c/go/+/230017
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Than McIntosh <thanm@google.... | [
{
"path": "src/cmd/link/internal/ld/util.go",
"patch": "@@ -74,7 +74,7 @@ func Errorf(s *sym.Symbol, format string, args ...interface{}) {\n // output file and return a non-zero error code.\n func (ctxt *Link) Errorf(s loader.Sym, format string, args ...interface{}) {\n \tif ctxt.loader != nil {\n-\t\tctxt.... | 2020-04-24T19:18:54 |
vuejs/vue | 02f897a4bc57fec89301918ed7c3db0c423deaae | d2902ca8ec5fd184fe81479fea1318553fdb8323 | chore: fix flow | [
{
"path": "src/core/vdom/create-functional-component.js",
"patch": "@@ -51,12 +51,12 @@ export function FunctionalRenderContext (\n this.injections = resolveInject(options.inject, parent)\n this.slots = () => resolveSlots(children, parent)\n \n- Object.defineProperty(this, 'scopedSlots', {\n+ Object.d... | 2019-01-09T22:41:37 |
ggml-org/llama.cpp | c959b676be29e93f8dbc3bd6056ceba812a9eb72 | cd08fc3ecc0264b4414b68af3874a6c689ed60c1 | CUDA: fix FA occupancy, optimize tile kernel (#15982) | [
{
"path": "ggml/src/ggml-cuda/common.cuh",
"patch": "@@ -75,6 +75,8 @@\n #define GGML_CUDA_CC_IS_RDNA4(cc) (cc >= GGML_CUDA_CC_RDNA4)\n #define GGML_CUDA_CC_IS_GCN(cc) (cc > GGML_CUDA_CC_OFFSET_AMD && cc < GGML_CUDA_CC_CDNA1)\n #define GGML_CUDA_CC_IS_CDNA(cc) (cc >= GGML_CUDA_CC_CDNA1 && cc < GGML_CUDA_... | 2025-09-17T13:32:42 |
huggingface/transformers | f9a98c476c3a92beaba3b20e51f1ff49417231a6 | 814619f54f677df79a337396794325f13f96251f | [`Mixtral` & `Mistral`] Add support for sdpa (#28133)
* some nits
* update test
* add support d\sd[a
* remove some dummy inputs
* all good
* style
* nits
* fixes
* fix more copies
* nits
* styling
* fix
* Update src/transformers/models/mistral/modeling_mistral.py
Co-authored-by: ... | [
{
"path": "src/transformers/models/mistral/modeling_mistral.py",
"patch": "@@ -31,7 +31,7 @@\n \n from ...activations import ACT2FN\n from ...cache_utils import Cache, DynamicCache\n-from ...modeling_attn_mask_utils import _prepare_4d_causal_attention_mask\n+from ...modeling_attn_mask_utils import _prepare_... | 2023-12-21T11:38:22 |
nodejs/node | 4796fbc7d820240dde544b89755285e3af5bcf6c | 472a66517a2a34fdce77cd0f6f29a2e62e7fd5d9 | doc: add docs for AssertionError
Fixes: https://github.com/nodejs/node/issues/12274
PR-URL: https://github.com/nodejs/node/pull/14261
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Tobias Nießen <tniessen... | [
{
"path": "doc/api/errors.md",
"patch": "@@ -358,6 +358,18 @@ loop tick.\n System-level errors are generated as augmented `Error` instances, which are\n detailed [here](#errors_system_errors).\n \n+## Class: AssertionError\n+\n+A subclass of `Error` that indicates the failure of an assertion. Such errors\n+... | 2017-07-16T06:02:04 |
vuejs/vue | d278da20a008a61ed8d5ec9d92524eb5dc9630cf | bbfeb039f37d8d040bde318fe904761c6bd9386f | test: fix LRU upgrade | [
{
"path": "test/ssr/ssr-bundle-render.spec.js",
"patch": "@@ -195,7 +195,7 @@ function createAssertions (runInNewContext) {\n })\n \n it('render with cache (nested)', done => {\n- const cache = LRU({ maxAge: Infinity })\n+ const cache = new LRU({ maxAge: Infinity })\n spyOn(cache, 'get').and.c... | 2019-01-07T17:15:15 |
golang/go | 1cc46d3a255ee0ecbb2b602095e6e4ec4d22fd61 | c53d1236dfaa900387519810c8de1a9353c55bab | runtime: sleep in TestSegv program to let signal be delivered
Since we're sleeping rather than waiting for the goroutines,
let the goroutines run forever.
Fixes #38595
Change-Id: I4cd611fd7565f6e8d91e50c9273d91c514825314
Reviewed-on: https://go-review.googlesource.com/c/go/+/229484
Run-TryBot: Ian Lance Taylor <iant... | [
{
"path": "src/runtime/testdata/testprogcgo/segv.go",
"patch": "@@ -10,8 +10,8 @@ package main\n import \"C\"\n \n import (\n-\t\"sync\"\n \t\"syscall\"\n+\t\"time\"\n )\n \n func init() {\n@@ -23,12 +23,9 @@ var Sum int\n \n func Segv() {\n \tc := make(chan bool)\n-\tvar wg sync.WaitGroup\n-\twg.Add(1)\n \... | 2020-04-22T23:03:30 |
ggml-org/llama.cpp | cd08fc3ecc0264b4414b68af3874a6c689ed60c1 | cb5bb6cc05119c24e7711ca2956cd0e6d409d396 | common : Fix corrupted memory error on json grammar initialization (#16038)
Initalizing RESERVED_NAME in is_reserved_name() is not thread
safe and leads to corrupted memory when used from multiple threads
as can be seen in the asan trace below. This fixes the initialization
to make it thread-safe.
#0 0x000100abd0... | [
{
"path": "common/json-schema-to-grammar.cpp",
"patch": "@@ -257,12 +257,13 @@ std::unordered_map<std::string, BuiltinRule> STRING_FORMAT_RULES = {\n };\n \n static bool is_reserved_name(const std::string & name) {\n- static std::unordered_set<std::string> RESERVED_NAMES;\n- if (RESERVED_NAMES.empty()... | 2025-09-17T08:08:02 |
huggingface/transformers | 7e93ce40c5c3cb60c1f73ec4c2cc6b4693d08599 | 4f7806ef7e6d717485a514c29fe60f09871a58f8 | Fix `input_embeds` docstring in encoder-decoder architectures (#28168) | [
{
"path": "src/transformers/models/bart/modeling_bart.py",
"patch": "@@ -1011,10 +1011,11 @@ def __init_subclass__(self):\n \n If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that\n don't have their past key value states given to ... | 2023-12-21T11:01:54 |
nodejs/node | c7a9a2edd426fc6e0113b4b21e11b042ba1f9aae | fe54bc7b6926fa51a39c431ffe350bd02faf53e9 | test: split out load-sensitive readline tests
Two test cases in `test-readline-interface` are sensitive to resource
constraints (probably due to `\r` and `\n` not arriving within the
appropriate delay to be treated as a single line ending). Move those
tests to `sequential`.
PR-URL: https://github.com/nodejs/node/pull... | [
{
"path": "test/parallel/test-readline-interface.js",
"patch": "@@ -22,6 +22,7 @@\n // Flags: --expose_internals\n 'use strict';\n const common = require('../common');\n+\n const assert = require('assert');\n const readline = require('readline');\n const internalReadline = require('internal/readline');\n@@ ... | 2017-08-08T00:17:31 |
golang/go | 512277dc19b918da19e083c28427109256ef9309 | da33f9c78a89c0997269a77e134127c135583963 | go/types: improve error message for pointer receiver errors
The compiler produces high quality error messages when an interface is
implemented by *T, rather than T. This change improves the analogous
error messages in go/types, from "missing method X" to "missing method
X (X has pointer receiver)".
I am open to impro... | [
{
"path": "src/go/types/expr.go",
"patch": "@@ -1566,10 +1566,13 @@ func (check *Checker) typeAssertion(pos token.Pos, x *operand, xtyp *Interface,\n \tif method == nil {\n \t\treturn\n \t}\n-\n \tvar msg string\n \tif wrongType != nil {\n-\t\tmsg = fmt.Sprintf(\"wrong type for method %s (have %s, want %s)\... | 2020-04-24T05:01:48 |
ggml-org/llama.cpp | 745cbcf2fe1eb88f8db615ac622f0b944d924ad6 | 1cbd80f8cf80a817715b1ccc5680fe2a3c5172c8 | llama-quant : fix the verification of attention layers for encoder-decoder models (#16023)
Signed-off-by: Jie Fu <jiefu@tencent.com> | [
{
"path": "src/llama-quant.cpp",
"patch": "@@ -725,7 +725,9 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::\n // attention layers have a non-zero number of kv heads\n int32_t n_attn_layer = model.hparams.n_layer - std::count(n_head_kv_iter, n_head_kv_iter ... | 2025-09-17T07:30:55 |
huggingface/transformers | 4f7806ef7e6d717485a514c29fe60f09871a58f8 | e268d7e5dcfc3a2f070114795592853ab4408473 | [bnb] Let's make serialization of 4bit models possible (#26037)
* updated bitsandbytes.py
* rm test_raise_* from test_4bit.py
* add test_4bit_serialization.py
* modeling_utils bulk edits
* bnb_ver 0.41.3 in integrations/bitsandbytes.py
Co-authored-by: Younes Belkada <49240599+younesbelkada@users.norepl... | [
{
"path": "src/transformers/integrations/bitsandbytes.py",
"patch": "@@ -21,7 +21,7 @@\n logger = logging.get_logger(__name__)\n \n \n-def set_module_quantized_tensor_to_device(module, tensor_name, device, value=None, fp16_statistics=None):\n+def set_module_quantized_tensor_to_device(module, tensor_name, de... | 2023-12-21T10:54:44 |
nodejs/node | 02371c7a8cbf150333638b8aaa9bb1da67164e3c | 06ed0e81879c97d6f9cbbbc0bf2b3db06ba232cf | doc: fix order of AtExit callbacks in addons.md
The sanity_check AtExit callback needs to come last to verify that the
other callbacks have been completed. This was not noticed before as this
code was not been executed.
PR-URL: https://github.com/nodejs/node/pull/14048
Reviewed-By: Michaël Zasso <targos@protonmail.co... | [
{
"path": "doc/api/addons.md",
"patch": "@@ -1109,10 +1109,10 @@ static void sanity_check(void*) {\n }\n \n void init(Local<Object> exports) {\n- AtExit(sanity_check);\n AtExit(at_exit_cb2, cookie);\n AtExit(at_exit_cb2, cookie);\n AtExit(at_exit_cb1, exports->GetIsolate());\n+ AtExit(sanity_check);... | 2017-07-03T05:56:43 |
ggml-org/llama.cpp | 85286f354813056f6c835046c0acfa3bf6ba9432 | d5fabe3682de515fd09d6c981f7a0d1b75614455 | model : add OLMo3 support (#16015)
* Add HF to gguf conversion logic for Olmo3
* Add Olmo3 implementation
* Update rope comment
* Fix indentation
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* Apply suggestion from @CISC
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
---------
... | [
{
"path": "convert_hf_to_gguf.py",
"patch": "@@ -6009,9 +6009,34 @@ class SeedOssModel(TextModel):\n \n \n @ModelBase.register(\"Olmo2ForCausalLM\")\n+@ModelBase.register(\"Olmo3ForCausalLM\")\n class Olmo2Model(TextModel):\n model_arch = gguf.MODEL_ARCH.OLMO2\n \n+ def set_gguf_parameters(self):\n+ ... | 2025-09-17T07:01:58 |
huggingface/transformers | 1d7773594754457ed4a79cf6d98bcaabea5bff51 | 45b70384a7d6692a8304f34a981a5ff020918b82 | Fix yolos resizing (#27663)
* Fix yolos resizing
* Update tests
* Add a test | [
{
"path": "src/transformers/models/detr/image_processing_detr.py",
"patch": "@@ -82,6 +82,7 @@\n SUPPORTED_ANNOTATION_FORMATS = (AnnotationFormat.COCO_DETECTION, AnnotationFormat.COCO_PANOPTIC)\n \n \n+# From the original repo: https://github.com/facebookresearch/detr/blob/3af9fa878e73b6894ce3596450a8d9b89d... | 2023-12-20T20:55:51 |
golang/go | e08f10b8b5fbb82ff1e2c263ad57e19d2de1e323 | 880ef2da7b81fe2e4e9fb75f4677377eeba70d1e | [dev.link] cmd/internal/goobj2: add index fingerprint to object file
The new object files use indices for symbol references, instead
of names. Fundamental to the design, it requires that the
importing and imported packages have consistent view of symbol
indices. The Go command should already ensure this, when using
"g... | [
{
"path": "src/cmd/compile/internal/gc/iexport.go",
"patch": "@@ -35,6 +35,8 @@\n // }\n // }\n //\n+// Fingerprint [8]byte\n+//\n // uvarint means a uint64 written out using uvarint encoding.\n //\n // []T means a uvarint followed by that many T objects. In other\n@@ -296,6 +298,10 @@ func ... | 2020-04-22T23:21:30 |
vuejs/vue | ce35159d6d6e9c3607bd3232e386b5b994d0ec49 | c7c13c2a156269d29fd9c9f8f6a3e53a2f2cac3d | chore: fix types | [
{
"path": "src/core/vdom/helpers/normalize-scoped-slots.js",
"patch": "@@ -2,7 +2,7 @@\n \n import { emptyObject } from 'core/util/index'\n \n-export function normalizeScopedSlots (slots: { [key: string]: Function }) {\n+export function normalizeScopedSlots (slots: { [key: string]: Function } | void): any {... | 2018-12-26T22:22:32 |
nodejs/node | 06ed0e81879c97d6f9cbbbc0bf2b3db06ba232cf | 7f81400723f602467e9f611cbef5addb31ced616 | doc: remove undef NDEBUG from addons.md
When working on commit 0d95a0b60a87533c54b517d132e16401fe28909b
("test: remove undef NDEBUG from at-exit addons test) I searched for
usages of undef NDEBUG but did not include the doc directory (but I did
include the test directory) and missed this one.
Commit 1f02569f8db9cb010... | [
{
"path": "doc/api/addons.md",
"patch": "@@ -1073,7 +1073,6 @@ The following `addon.cc` implements AtExit:\n \n ```cpp\n // addon.cc\n-#undef NDEBUG\n #include <assert.h>\n #include <stdlib.h>\n #include <node.h>",
"additions": 0,
"deletions": 1,
"language": "Markdown"
}
] | 2017-07-03T04:58:45 |
ggml-org/llama.cpp | d5fabe3682de515fd09d6c981f7a0d1b75614455 | 8ff206097c2bf3ca1c7aa95f9d6db779fc7bdd68 | CANN: Optimize ggml_cann_set_device (#15935)
* CANN: Fix ggml_cann_set_device to avoid redundant device switches
- Added a check to skip aclrtSetDevice if the current device is already set.
- Prevents unnecessary context switches while keeping thread/device consistency.
* CANN: add device default id | [
{
"path": "ggml/src/ggml-cann/common.h",
"patch": "@@ -526,7 +526,10 @@ struct ggml_backend_cann_context {\n */\n aclrtStream stream(int stream) {\n if (streams[stream] == nullptr) {\n- ggml_cann_set_device(device);\n+ // If the device is not set here, destroying the s... | 2025-09-17T06:33:08 |
huggingface/transformers | 45b70384a7d6692a8304f34a981a5ff020918b82 | 01c081d1381e20b1f5b2f31d6fa8b0af8092c0b5 | Generate: fix speculative decoding (#28166)
Co-authored-by: Merve Noyan <merveenoyan@gmail.com> | [
{
"path": "docs/source/en/generation_strategies.md",
"patch": "@@ -82,7 +82,7 @@ Even if the default decoding strategy mostly works for your task, you can still\n commonly adjusted parameters include:\n \n - `max_new_tokens`: the maximum number of tokens to generate. In other words, the size of the output s... | 2023-12-20T18:55:35 |
vuejs/vue | c7c13c2a156269d29fd9c9f8f6a3e53a2f2cac3d | d747469e1ae26190d364e5cc9c8112d3d865a0b9 | fix(scoped-slots): ensure $scopedSlots calls always return Arrays
Also allow render functions to return an Array of a single element.
Close #8056 | [
{
"path": "src/core/instance/render-helpers/resolve-slots.js",
"patch": "@@ -55,10 +55,11 @@ export function resolveScopedSlots (\n ): { [key: string]: Function } {\n res = res || {}\n for (let i = 0; i < fns.length; i++) {\n- if (Array.isArray(fns[i])) {\n- resolveScopedSlots(fns[i], res)\n+ ... | 2018-12-26T22:17:30 |
golang/go | 880ef2da7b81fe2e4e9fb75f4677377eeba70d1e | f2d8da1a353b493548e211b06fdf47bd998ad4b6 | [dev.link] cmd/link: panic if HeadType is not set
In the code there are conditions like !ctxt.IsDarwin(). This will
accidentally be true if HeadType is not yet set. Panic when
HeadType is not set, to catch errors.
Change-Id: Ic891123f27f0276fff5a4b5d29e5b1f7ebbb94ee
Reviewed-on: https://go-review.googlesource.com/c/g... | [
{
"path": "src/cmd/link/internal/ld/target.go",
"patch": "@@ -61,6 +61,7 @@ func (t *Target) CanUsePlugins() bool {\n }\n \n func (t *Target) IsElf() bool {\n+\tt.mustSetHeadType()\n \treturn t.IsELF\n }\n \n@@ -112,37 +113,51 @@ func (t *Target) IsWasm() bool {\n //\n \n func (t *Target) IsLinux() bool {\n... | 2020-04-24T17:02:37 |
nodejs/node | 7f81400723f602467e9f611cbef5addb31ced616 | a439cf4354c4781c7ac0026c7f69f7160e8439fb | tools: fix tools/addon-verify.js
The current implementation of addon-verify.js is including the code
for the "Function arguments" section in test/addons/01_callbacks and
there is no directory generated or the "Function arguments section".
This continues and leads to the last section, "AtExit", code to be
excluded. The... | [
{
"path": "doc/api/addons.md",
"patch": "@@ -1125,7 +1125,7 @@ Test in JavaScript by running:\n \n ```js\n // test.js\n-const addon = require('./build/Release/addon');\n+require('./build/Release/addon');\n ```\n \n [Embedder's Guide]: https://github.com/v8/v8/wiki/Embedder's%20Guide",
"additions": 1,
... | 2017-07-03T04:27:58 |
ggml-org/llama.cpp | 3913f8730ec6d6245480affc30ae3049107956f4 | 76888d202ed2b835ae19ea9f9db6baf39e419297 | ggml : fix padding in timestep embedding kernels (#15932)
* ggml : remove adding extra dim timestep embedding
This commit updates the ggml_timestep_embedding function to no longer
add an extra dimension when the specified dimension is odd.
The motivation for this change is that this introduces an unnecessary
dimensi... | [
{
"path": "ggml/src/ggml-cpu/ops.cpp",
"patch": "@@ -8599,7 +8599,6 @@ static void ggml_compute_forward_timestep_embedding_f32(\n }\n if (dim % 2 != 0 && ith == 0) {\n embed_data[2 * half] = 0.f;\n- embed_data[dim] = 0.f;\n }\n }\n }",
"additions": 0,
... | 2025-09-16T13:25:57 |
huggingface/transformers | 01c081d1381e20b1f5b2f31d6fa8b0af8092c0b5 | ee298a16a262a6b3b12b8b15232e76516bb562fd | [docs] Trainer docs (#28145)
* fsdp, debugging, gpu selection
* fix hfoption
* fix | [
{
"path": "docs/source/en/_toctree.yml",
"patch": "@@ -111,6 +111,8 @@\n title: Share a custom model\n - local: chat_templating\n title: Templates for chat models\n+ - local: trainer\n+ title: Trainer\n - local: sagemaker\n title: Run training on Amazon SageMaker\n - local: serializati... | 2023-12-20T18:37:23 |
vuejs/vue | 7a9290421d511cca66a6563f910beb1fab1b33ee | 85d8ad64aa864f4c0800fbb9a895432b5edfdfe3 | build: fix chrome headless on CI | [
{
"path": "test/unit/karma.cover.config.js",
"patch": "@@ -4,7 +4,13 @@ process.env.CHROME_BIN = require('puppeteer').executablePath()\n \n module.exports = function (config) {\n const options = Object.assign(base, {\n- browsers: ['ChromeHeadless'],\n+ browsers: ['ChromeHeadlessCI'],\n+ customLau... | 2018-12-26T20:48:19 |
golang/go | 3d1007d28e92316b08974dc65cde44a6aa31bf21 | 442fd182fba66a11f9cbe3321217e8b4618fcfc7 | [dev.link] cmd/link: move more error handling into loader
Move the guts of ctxt.Errorf into loader.Loader, so that we can make
calls to it from functions that have a "*loader.Loader" available but
not a "ctxt *Link". This is needed to start converting hooks like
"adddynrel" in the arch-specific portions of the linker ... | [
{
"path": "src/cmd/link/internal/ld/errors.go",
"patch": "@@ -7,8 +7,6 @@ import (\n \t\"cmd/internal/obj\"\n \t\"cmd/link/internal/loader\"\n \t\"cmd/link/internal/sym\"\n-\t\"fmt\"\n-\t\"os\"\n \t\"sync\"\n )\n \n@@ -22,6 +20,7 @@ type symNameFn func(s loader.Sym) string\n \n // ErrorReporter is used to m... | 2020-04-22T19:16:06 |
ggml-org/llama.cpp | f1fbffb5c0b34b2a68febb7da3fd0f8333f1ed4c | 51abc96bdc52ba8cd6ad78dcf12ed9a041d7b442 | fix: apply clang-format to CUDA macros (#16017)
clang-format previously broke long CUDA macros (e.g. __launch_bounds__) into
unreadable line breaks inside template declarations, such as:
template<int D, int ncols, int nwarps, int VKQ_stride,
typename KQ_acc_t, bool use_logit_softcap>
__launch_bound... | [
{
"path": ".clang-format",
"patch": "@@ -22,6 +22,13 @@ AllowShortIfStatementsOnASingleLine: Never\n AllowShortLambdasOnASingleLine: Inline\n AllowShortLoopsOnASingleLine: false\n AlwaysBreakBeforeMultilineStrings: true\n+# Treat CUDA keywords/attributes as \"attribute macros\" and avoid breaking lines insi... | 2025-09-16T06:59:19 |
rust-lang/rust | 3a041fcc2cc514ef87f29d8aa8300a0f6a2fa0da | da4c09937559e8d93cc6499ec8510a3571449061 | Propogate error types in mir type projections | [
{
"path": "src/tools/rust-analyzer/crates/hir-ty/src/mir.rs",
"patch": "@@ -10,7 +10,7 @@ use crate::{\n lang_items::is_box,\n mapping::ToChalk,\n CallableDefId, ClosureId, Const, ConstScalar, InferenceResult, Interner, MemoryMap,\n- Substitution, TraitEnvironment, Ty, TyKind,\n+ Substitut... | 2025-02-12T08:24:33 |
huggingface/transformers | ee298a16a262a6b3b12b8b15232e76516bb562fd | 224ab70969d1ac6c549f0beb3a8a71e2222e50f7 | Align backbone stage selection with out_indices & out_features (#27606)
* Iteratre over out_features instead of stage_names
* Update for all backbones
* Add tests
* Fix
* Align timm backbone behaviour with other backbones
* Fix tests
* Stricter checks on set out_features and out_indices
* Revert b... | [
{
"path": "src/transformers/models/beit/configuration_beit.py",
"patch": "@@ -102,11 +102,13 @@ class BeitConfig(BackboneConfigMixin, PretrainedConfig):\n out_features (`List[str]`, *optional*):\n If used as backbone, list of features to output. Can be any of `\"stem\"`, `\"stage1\"`, `\... | 2023-12-20T18:33:17 |
vuejs/vue | b31a1aa8870474b2ca782c45d55edac2932d4cc2 | d08b49f520e0704f9d4e61be4f751e3b2cdac6a8 | feat(compiler): output source range for compiler errors (#7127)
ref #6338 | [
{
"path": "flow/compiler.js",
"patch": "@@ -18,6 +18,7 @@ declare type CompilerOptions = {\n shouldDecodeTags?: boolean;\n shouldDecodeNewlines?: boolean;\n shouldDecodeNewlinesForHref?: boolean;\n+ outputSourceRange?: boolean;\n \n // runtime user-configurable\n delimiters?: [string, string]; /... | 2018-12-22T02:58:32 |
nodejs/node | a439cf4354c4781c7ac0026c7f69f7160e8439fb | b72e702247a4dcbbb9d678253dad2b9b223459f4 | test: fix conversion of microseconds in test
PR-URL: https://github.com/nodejs/node/pull/14706
Fixes: https://github.com/nodejs/node/issues/8728
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Colin Ihrig <cjihrig@gma... | [
{
"path": "test/pummel/test-process-cpuUsage.js",
"patch": "@@ -17,14 +17,14 @@ while (Date.now() - now < RUN_FOR_MS);\n // Get a diff reading from when we started.\n const diff = process.cpuUsage(start);\n \n-const MICROSECONDS_PER_SECOND = 1000 * 1000;\n+const MICROSECONDS_PER_MILLISECOND = 1000;\n \n // ... | 2017-08-09T03:48:47 |
golang/go | adea6a90e361629d20a68400c0c5cdcdfcdf087e | 67bf856b962731965f7e9a75e6bede62fee1a4da | [dev.link] cmd/link/internal/loader: fix buglet in section handling
Allow for the possibility that a client could call newExtSym(), then
ask for the section of the new sym before SetSectSym is called on it
(check in SymSect for this case).
Change-Id: I7bd78e7b3b7618943705b616f62ea78c4a1b68d0
Reviewed-on: https://go-r... | [
{
"path": "src/cmd/link/internal/loader/loader.go",
"patch": "@@ -1075,6 +1075,12 @@ func (l *Loader) SetSymAlign(i Sym, align int32) {\n \n // SymValue returns the section of the i-th symbol. i is global index.\n func (l *Loader) SymSect(i Sym) *sym.Section {\n+\tif int(i) >= len(l.symSects) {\n+\t\t// sym... | 2020-04-23T12:43:29 |
ggml-org/llama.cpp | 3d4053f77f0f78ee2b791088c02af653ebee42dd | dc381aa9a6dc45f00673471d34b8bddd30e77570 | CUDA: fix im2col_3d to respect non-contiguous inputs (views) (#15956)
* fix im2col_3d to respect non-contiguous inputs (views)
The CUDA 3D im2col kernel computed source addresses assuming compact layout (products of dims), ignoring nb[] strides.
This patch switches im2col_3d source indexing to use true strides de... | [
{
"path": "ggml/src/ggml-cuda/im2col.cu",
"patch": "@@ -122,11 +122,14 @@ static __global__ void im2col_3d_kernel(\n int64_t OH_OW, int64_t KD_KH_KW, int64_t ID_IH_IW, int64_t KH_KW, int64_t IH_IW, int64_t IC_ID_IH_IW,\n int64_t IC_KD_KH_KW, int64_t OW_KD_KH_KW, int64_t OD_OH_OW_IC_KD_KH_KW... | 2025-09-15T22:28:31 |
vuejs/vue | d08b49f520e0704f9d4e61be4f751e3b2cdac6a8 | 1c105fb16c1b2395790740efce10b137f2f36c19 | fix: make transition-group key warning a tip to avoid breaking compilation | [
{
"path": "src/compiler/parser/index.js",
"patch": "@@ -341,7 +341,8 @@ function processKey (el) {\n if (iterator && iterator === exp && parent && parent.tag === 'transition-group') {\n warn(\n `Do not use v-for index as key on <transition-group> children, ` +\n- `th... | 2018-12-21T19:46:18 |
huggingface/transformers | 224ab70969d1ac6c549f0beb3a8a71e2222e50f7 | 9924df9eb234b979ee3fb3f5463e039b9e9623dd | Update FA2 exception msg to point to hub discussions (#28161)
* Update FA2 exception msg to point to hub discussions
* Use path for hub url | [
{
"path": "src/transformers/modeling_utils.py",
"patch": "@@ -1380,8 +1380,9 @@ def _check_and_enable_flash_attn_2(\n \"\"\"\n if not cls._supports_flash_attn_2:\n raise ValueError(\n- f\"{cls.__name__} does not support Flash Attention 2.0 yet. Please open an issue... | 2023-12-20T16:52:16 |
nodejs/node | b72e702247a4dcbbb9d678253dad2b9b223459f4 | 2923ed10bb6dd1d6a203e1a89d1aa179df5cf243 | 2017-08-09, Version 8.3.0 (Current)
V8 6.0:
The V8 engine has been upgraded to version 6.0, which has a significantly
changed performance profile.
[#14574](https://github.com/nodejs/node/pull/14574)
More detailed information on performance differences can be found at
https://medium.com/the-node-js-collecti... | [
{
"path": "CHANGELOG.md",
"patch": "@@ -27,7 +27,8 @@ release.\n </tr>\n <tr>\n <td valign=\"top\">\n-<b><a href=\"doc/changelogs/CHANGELOG_V8.md#8.2.1\">8.2.1</a></b><br/>\n+<b><a href=\"doc/changelogs/CHANGELOG_V8.md#8.3.0\">8.3.0</a></b><br/>\n+<a href=\"doc/changelogs/CHANGELOG_V8.md#8.2.1\">8.2.1</... | 2017-08-02T10:52:34 |
golang/go | d0ea533c54092e034ab474691df56f8da05ed0d5 | f5fcc9b8e08aa94c1f8ad2b91f447e32940a5435 | cmd/compile: fix misalignment in sources column of generated ssa.html
Fix regression where line numbers in the sources column of generated
ssa.html output became misaligned with the source code. This was due
to some new margins applied to certain h2 elements during the work
to combine identical columns.
Fixes #38612
... | [
{
"path": "src/cmd/compile/internal/ssa/html.go",
"patch": "@@ -143,6 +143,7 @@ pre {\n float: left;\n overflow: hidden;\n text-align: right;\n+ margin-top: 7px;\n }\n \n .lines div {",
"additions": 1,
"deletions": 0,
"language": "Go"
}
] | 2020-04-24T00:46:43 |
ggml-org/llama.cpp | 28c39da7c645185ade5436767929d7ec33006033 | 106220562aca42b6738b8f51acfce0db1b8a2fb6 | llama-run: Fix model download on Windows (#15988)
* llama-run: Fix model download on Windows
* fix SSL error (SSL peer certificate or SSH remote key was not OK)
* fix program crash on std::filesystem::rename
* llama-run: create a separate method to utilize RAII
* llama-run: handle rename exception | [
{
"path": "tools/run/run.cpp",
"patch": "@@ -407,6 +407,43 @@ class HttpClient {\n }\n \n std::string output_file_partial;\n+\n+ if (!output_file.empty()) {\n+ output_file_partial = output_file + \".partial\";\n+ }\n+\n+ if (download(url, headers, output_file_... | 2025-09-15T10:08:30 |
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.