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 |
|---|---|---|---|---|---|
golang/go | a29dd1f2bd0ee0048f03782ba67034a0441a147e | cb768c801a9cafa18fe7805ac70bd947930fcdc7 | net: add Example for Unix dialer with context
Showing users how to accomplish this more easily is a compromise
between doing nothing and adding a new API to the net package.
Fixes #38506.
Change-Id: I43f831cf94951c987cf3c8c1aa55f0012ee8034e
Reviewed-on: https://go-review.googlesource.com/c/go/+/263417
Reviewed-by: I... | [
{
"path": "src/net/example_test.go",
"patch": "@@ -55,6 +55,27 @@ func ExampleDialer() {\n \t}\n }\n \n+func ExampleDialer_unix() {\n+\t// DialUnix does not take a context.Context parameter. This example shows\n+\t// how to dial a Unix socket with a Context. Note that the Context only\n+\t// applies to the ... | 2020-10-18T16:49:16 |
ollama/ollama | e1f1c374ea3033d55e4975fbb60a5873f716b8ba | 06a1508bfe456e82ba053ea554264e140c5057b5 | Community Integration: ChatOllama (#3400)
* Community Integration: ChatOllama
* fixed typo | [
{
"path": "README.md",
"patch": "@@ -289,6 +289,7 @@ See the [API documentation](./docs/api.md) for all endpoints.\n - [AnythingLLM (Docker + MacOs/Windows/Linux native app)](https://github.com/Mintplex-Labs/anything-llm)\n - [Ollama Basic Chat: Uses HyperDiv Reactive UI](https://github.com/rapidarchitect/o... | 2024-03-31T02:46:50 |
ggml-org/llama.cpp | 59db9a357d9a247009c70fda34050661b17a1a5c | 23fbfcb1ad6c6f76b230e8895254de785000be46 | llama: dynamic head_dim and n_rot for SWA (#20301)
* llama: dynamic head_dim and n_rot for SWA
* also add gguf_writer wrappers
* fix build
* build_rope_shift arg reorder | [
{
"path": "gguf-py/gguf/constants.py",
"patch": "@@ -177,6 +177,8 @@ class Attention:\n TEMPERATURE_LENGTH = \"{arch}.attention.temperature_length\"\n KEY_LENGTH_MLA = \"{arch}.attention.key_length_mla\"\n VALUE_LENGTH_MLA = \"{arch}.attention.valu... | 2026-03-09T21:22:39 |
huggingface/transformers | dc6eb448419b6a4a4967cdc7e8b96bdbc6969644 | 517df566f572d90e6301df87870f651f0d1b1110 | Improve error msg when using bitsandbytes (#31350)
improve error msg when using bnb | [
{
"path": "src/transformers/quantizers/quantizer_bnb_4bit.py",
"patch": "@@ -60,10 +60,11 @@ def __init__(self, quantization_config, **kwargs):\n def validate_environment(self, *args, **kwargs):\n if not torch.cuda.is_available():\n raise RuntimeError(\"No GPU found. A GPU is needed ... | 2024-06-10T12:22:14 |
nodejs/node | 7b3446e7bb19b2df45892300a13c5bbf30da244c | 941c65bba28ed22c09cb262d04512fc7fecab189 | test: unmark flaky test
This effectively reverts 5c475a722357b1d4ddf099992e3f5df7e51583e1 and
75095d700137e29ad1413972b8f6eed7c3473a65.
PR-URL: https://github.com/nodejs/node/pull/16758
Fixes: https://github.com/node/issues/16210
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.c... | [
{
"path": "test/parallel/parallel.status",
"patch": "@@ -5,8 +5,6 @@ prefix parallel\n # sample-test : PASS,FLAKY\n \n [true] # This section applies to all platforms\n-# https://github.com/nodejs/node/issues/16210\n-test-async-wrap-uncaughtexception: PASS,FLAKY\n \n [$system==win32]\n... | 2017-11-04T21:36:18 |
golang/go | cb768c801a9cafa18fe7805ac70bd947930fcdc7 | 9a49f772575f8009355ac7d64d51ef41bc960c3d | encoding/xml: fix reserved namespace check to be case-insensitive
Fixes the check for the reserved namespace prefix
"xml" to be case insensitive, so as to match all variants of:
(('X'|'x')('M'|'m')('L'|'l'))
as mandated by Section 2.3 of https://www.w3.org/TR/REC-xml/
This is a roll forward of CL 203417, which ... | [
{
"path": "doc/go1.16.html",
"patch": "@@ -302,6 +302,20 @@ <h3 id=\"minor_library_changes\">Minor changes to the library</h3>\n </dd>\n </dl><!-- crypto/x509 -->\n \n+<dl id=\"encoding/xml\"><dt><a href=\"/pkg/encoding/xml/\">encoding/xml</a></dt>\n+ <dd>\n+ <p><!-- CL 264024 -->\n+ The encoder ... | 2020-10-20T22:19:36 |
rust-lang/rust | 5339d4ef5b3e15a92594dd397c38db036566f54f | ef031c854da6432c86ac14c438ab96c7e349d85f | Fix target-feature inline test to be less flaky | [
{
"path": "tests/codegen/target-feature-inline-closure.rs",
"patch": "@@ -12,7 +12,7 @@ use std::arch::x86_64::*;\n #[cfg(target_arch = \"x86_64\")]\n #[target_feature(enable = \"avx\")]\n fn with_avx(x: __m256) -> __m256 {\n- // CHECK: fadd\n+ // CHECK: fadd <8 x float>\n let add = {\n #[... | 2025-03-06T19:56:21 |
ggml-org/llama.cpp | 23fbfcb1ad6c6f76b230e8895254de785000be46 | e22cd0aa155f0188eb9c605fd54daf3f2f38177b | server: Parse port numbers from MCP server URLs in CORS proxy (#20208)
* Parse port numbers from MCP server URLs
* Pass scheme to http proxy for determining whether to use SSL
* Fix download on non-standard port and re-add port to logging
* add test
---------
Co-authored-by: Xuan Son Nguyen <son@huggingface.co> | [
{
"path": "common/http.h",
"patch": "@@ -7,6 +7,7 @@ struct common_http_url {\n std::string user;\n std::string password;\n std::string host;\n+ int port;\n std::string path;\n };\n \n@@ -47,6 +48,20 @@ static common_http_url common_http_parse_url(const std::string & url) {\n part... | 2026-03-09T16:47:54 |
ollama/ollama | 06a1508bfe456e82ba053ea554264e140c5057b5 | 5a5efee46b2b633ea0e8d69311b57ab4c2d71abe | Update 90_bug_report.yml | [
{
"path": ".github/ISSUE_TEMPLATE/90_bug_report.yml",
"patch": "@@ -19,7 +19,7 @@ body:\n label: What did you expect to see?\n description: What did you expect to see/happen instead?\n validations:\n- required: true\n+ required: false\n - type: textarea\n id: steps\n attr... | 2024-03-29T14:11:17 |
nodejs/node | 941c65bba28ed22c09cb262d04512fc7fecab189 | ff21851816f8512a1488265a878140d8aa67683a | src: use unrefed async for GC tracking
Do not let an internal handle keep the event loop alive.
PR-URL: https://github.com/nodejs/node/pull/16758
Fixes: https://github.com/node/issues/16210
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gm... | [
{
"path": "src/node_perf.cc",
"patch": "@@ -215,6 +215,7 @@ void MarkGarbageCollectionEnd(Isolate* isolate,\n uv_async_t* async = new uv_async_t(); // coverity[leaked_storage]\n if (uv_async_init(env->event_loop(), async, PerformanceGCCallback))\n return delete async;\n+ uv_unref(reinterpret_cast<... | 2017-11-04T21:25:18 |
golang/go | 9a49f772575f8009355ac7d64d51ef41bc960c3d | 4e5a313524da62600eb59dbf98624cfe946456f8 | runtime/race: update race .syso files
Fixes #39186
Change-Id: I624ab73b3083f190978c09716672ce1b712a5c81
Reviewed-on: https://go-review.googlesource.com/c/go/+/264082
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Keith Randall... | [
{
"path": "src/runtime/race/README",
"patch": "@@ -4,10 +4,10 @@ the LLVM project (https://github.com/llvm/llvm-project/tree/master/compiler-rt).\n \n To update the .syso files use golang.org/x/build/cmd/racebuild.\n \n-race_darwin_amd64.syso built with LLVM 3496d6e4bea9cb99cb382939b7e79a50a3b863a5 and Go 5... | 2020-10-21T03:08:30 |
huggingface/transformers | 517df566f572d90e6301df87870f651f0d1b1110 | 4fa4dcb2be1dd3d71a6a5fee14039d2d342520d9 | Decorators for deprecation and named arguments validation (#30799)
* Fix do_reduce_labels for maskformer image processor
* Deprecate reduce_labels in favor to do_reduce_labels
* Deprecate reduce_labels in favor to do_reduce_labels (segformer)
* Deprecate reduce_labels in favor to do_reduce_labels (oneformer)
... | [
{
"path": "docs/source/en/model_doc/segformer.md",
"patch": "@@ -66,12 +66,12 @@ of the model was contributed by [sayakpaul](https://huggingface.co/sayakpaul). T\n important preprocessing step is that images and segmentation maps are randomly cropped and padded to the same size,\n such as 512x512 or 640... | 2024-06-10T11:35:10 |
ggml-org/llama.cpp | 96cfc4992c625984d99f3529d30bedab1e1ecf93 | ed0007aa32b94f40e4f3ba0d0fe24d86a582232b | server : fix checkpoints n_tokens calculation (#20287) | [
{
"path": "tools/server/server-context.cpp",
"patch": "@@ -2141,6 +2141,9 @@ struct server_context_impl {\n if (slot.state == SLOT_STATE_PROCESSING_PROMPT || slot.state == SLOT_STATE_STARTED) {\n const auto & input_tokens = slot.task->tokens;\n \n+ // u... | 2026-03-09T14:47:06 |
rust-lang/rust | fcb34b17e22268636a7b227ce78237b158737d39 | 860637abc36c5ab3cdc1cf6a5b4130a2d8fecd03 | Pass the target crate in `HirFormatter`
This is required to format evaluated consts, because we need trait env, and it needs the crate (currently it uses the last crate in topological order, which is wrong, the next commit will fix that). | [
{
"path": "src/tools/rust-analyzer/crates/hir-ty/src/consteval.rs",
"patch": "@@ -15,9 +15,10 @@ use stdx::never;\n use triomphe::Arc;\n \n use crate::{\n- db::HirDatabase, generics::Generics, infer::InferenceContext, lower::ParamLoweringMode,\n- mir::monomorphize_mir_body_bad, to_placeholder_idx, Con... | 2025-02-27T12:19:54 |
nodejs/node | aa81f995b4bcbdc4bfb40b72e31058d40f5b7f3a | 694efa882ff32f77fe723e987e83e162c118ae82 | test: fix flaky test-http2-create-client-connect
The first group of tests makes one more connection and leave the server
alive for longer. Otherwise the test is just catching that the server
has closed the socket, depending on timing.
This does not quite make the test pass yet, however. There are some
quirks with how... | [
{
"path": "test/parallel/test-http2-create-client-connect.js",
"patch": "@@ -3,6 +3,7 @@\n // Tests http2.connect()\n \n const common = require('../common');\n+const Countdown = require('../common/countdown');\n if (!common.hasCrypto)\n common.skip('missing crypto');\n const fixtures = require('../common/... | 2017-09-23T17:07:38 |
ggml-org/llama.cpp | d6e1556499814da42424e39397a9964a1bebbf00 | f76565db92272d98976d5b8b1809ebe1e079f8cd | server : fix off-by-1 in server_tokens::size_up_to_pos() (#20279)
* server : fix off-by-1 in server_tokens::size_up_to_pos()
* cont : fix typo [no ci] | [
{
"path": "tools/server/server-common.cpp",
"patch": "@@ -276,7 +276,7 @@ llama_pos server_tokens::pos_next(int64_t n_tokens) const {\n \n size_t server_tokens::size_up_to_pos(llama_pos max_pos) const {\n if (!has_mtmd) {\n- return std::min((size_t)(max_pos + 1), tokens.size());\n+ return ... | 2026-03-09T14:43:38 |
huggingface/transformers | 4fa4dcb2be1dd3d71a6a5fee14039d2d342520d9 | 6b11f89c6b3d15655a2264df73e453cf9692c62e | docs/zh: fix style (#31334) | [
{
"path": "docs/source/zh/task_summary.md",
"patch": "@@ -284,7 +284,6 @@ score: 0.9327, start: 30, end: 54, answer: huggingface/transformers\n 有两种类型的话语模型：\n \n * causal：模型的目标是预测序列中的下一个`token`，而未来的`tokens`被遮盖。\n- \n \n ```py\n >>> from transformers import pipeline\n@@ -294,9 +293,8 @@ score: 0.9327... | 2024-06-10T10:40:40 |
golang/go | 4e5a313524da62600eb59dbf98624cfe946456f8 | b2a8317b31d652b3ee293a313269b8290bcdf96c | net/http: test that ParseMultipartForm catches overflows
Tests that if the combination of:
* HTTP multipart file payload size
* ParseMultipartForm's maxMemory parameter
* the internal leeway buffer size of 10MiB
overflows, then we'll report an overflow instead of silently
passing.
Reapplies and fixes CL 254977, whic... | [
{
"path": "src/net/http/request_test.go",
"patch": "@@ -13,6 +13,7 @@ import (\n \t\"fmt\"\n \t\"io\"\n \t\"io/ioutil\"\n+\t\"math\"\n \t\"mime/multipart\"\n \t. \"net/http\"\n \t\"net/http/httptest\"\n@@ -245,6 +246,50 @@ func TestParseMultipartForm(t *testing.T) {\n \t}\n }\n \n+// Issue #40430: Test that... | 2020-10-20T11:11:12 |
rust-lang/rust | ac40ea7129447abbd004859957bf00a374e8a739 | 30f168ef811aec63124eac677e14699baa9395bd | Suggest typo fix for static lifetime | [
{
"path": "compiler/rustc_resolve/src/late/diagnostics.rs",
"patch": "@@ -24,7 +24,7 @@ use rustc_hir::def_id::{CRATE_DEF_ID, DefId};\n use rustc_hir::{MissingLifetimeKind, PrimTy};\n use rustc_middle::ty;\n use rustc_session::{Session, lint};\n-use rustc_span::edit_distance::find_best_match_for_name;\n+use... | 2025-03-06T17:32:32 |
ggml-org/llama.cpp | 43e1cbd6c1b407fcb1fb0196276265e774986035 | 107d5999520dd02195ebe05278752db9fd33c865 | models : fix assert in mamba2 graph (#20270) | [
{
"path": "src/models/mamba-base.cpp",
"patch": "@@ -155,7 +155,6 @@ ggml_tensor * llm_build_mamba_base::build_mamba2_layer(llm_graph_input_rs * inp,\n \n const auto kv_head = mctx_cur->get_head();\n \n- const int64_t n_embd = hparams.n_embd;\n const int64_t d_conv = hparams.ssm_d_conv;\n ... | 2026-03-09T11:15:15 |
nodejs/node | 694efa882ff32f77fe723e987e83e162c118ae82 | 63c278959ad824e953c8703b94f534dd77a8c297 | crypto: clear some SSL_METHOD deprecation warnings
Fixing the rest will be rather involved. I think the cleanest option is
to deprecate the method string APIs which are weird to begin with.
PR-URL: https://github.com/nodejs/node/pull/16130
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Rod Vagg <rod@vag... | [
{
"path": "src/node_crypto.cc",
"patch": "@@ -178,6 +178,8 @@ static int DH_set0_key(DH* dh, BIGNUM* pub_key, BIGNUM* priv_key) {\n return 1;\n }\n \n+static const SSL_METHOD* TLS_method() { return SSLv23_method(); }\n+\n static void SSL_SESSION_get0_ticket(const SSL_SESSION* s,\n ... | 2017-09-18T16:34:04 |
huggingface/transformers | 6b11f89c6b3d15655a2264df73e453cf9692c62e | 807483edba45ee9707e55a36c572f8e2c3cd347e | Fix paligemma inverted mask (#31207)
* pass inverted causal mask
* add sanity check for paligemma finetuning
* [run-slow]paligemma | [
{
"path": "src/transformers/models/paligemma/modeling_paligemma.py",
"patch": "@@ -348,8 +348,11 @@ def _merge_input_ids_with_image_features(\n final_labels = torch.where(input_ids != self.pad_token_id, labels, final_labels)\n else:\n causal_mask = attention_mask.unsqueeze(1)... | 2024-06-10T09:22:39 |
golang/go | b2a8317b31d652b3ee293a313269b8290bcdf96c | 1c0d1f4e975eb864e36b18e63bcb8cdedbdfc51b | cmd/compile: use desired info when allocating registers for live values
When allocting registers for live values, use desired register if available,
this is helpful for some cases, such as (*entry).delete, which can save a
few of copies.
Besides, this patch allows more debugging information to be printed out.
Test re... | [
{
"path": "src/cmd/compile/internal/ssa/regalloc.go",
"patch": "@@ -1582,6 +1582,19 @@ func (s *regAllocState) regalloc(f *Func) {\n \t\t\t\t}\n \t\t\t\tv := s.orig[vid]\n \t\t\t\tm := s.compatRegs(v.Type) &^ s.used\n+\t\t\t\t// Used desired register if available.\n+\t\t\touterloop:\n+\t\t\t\tfor _, e := ra... | 2020-10-19T03:57:15 |
ollama/ollama | 5255d0af8ad66f2919b3abcd2afcae7a95d2cdca | 461ad25015cc9b760a49bdb8e8fd26c87c2a359c | fix: workflows | [
{
"path": ".github/workflows/test.yaml",
"patch": "@@ -15,10 +15,13 @@ jobs:\n GENERATE_CUDA: ${{ steps.changes.outputs.GENERATE_CUDA }}\n GENERATE_ROCM: ${{ steps.changes.outputs.GENERATE_ROCM }}\n steps:\n+ - uses: actions/checkout@v4\n+ with:\n+ fetch-depth: 0\n ... | 2024-03-27T23:14:20 |
huggingface/transformers | 807483edba45ee9707e55a36c572f8e2c3cd347e | 2f16a45d5f5328bf6c6f8687a7d3c4d5b2c8af97 | docs: fix style (#31340) | [
{
"path": "docs/source/en/perf_torch_compile.md",
"patch": "@@ -314,7 +314,7 @@ We also benchmarked on PyTorch nightly (2.1.0dev, find the wheel [here](https://\n | Object Detection/DETR | 4 | 269.615 | 204.785 |\n | Object Detection/DETR | 16 | OOM | OOM |\n \n-### V100\n+### V100\n \n | **Task/Model** |... | 2024-06-10T08:53:25 |
nodejs/node | 63c278959ad824e953c8703b94f534dd77a8c297 | d9b9229d98afb4b3f01e73c619c5fede0f87138f | crypto: make ALPN the same for OpenSSL 1.0.2 & 1.1.0
This is kind of hairy. OpenSSL 1.0.2 ignored the return value and always
treated everything as SSL_TLSEXT_ERR_NOACK (so the comment was wrong and
Node was never sending a warning alert). OpenSSL 1.1.0 honors
SSL_TLSEXT_ERR_NOACK vs SSL_TLSEXT_ERR_FATAL_ALERT and tre... | [
{
"path": "src/node_crypto.cc",
"patch": "@@ -2502,20 +2502,12 @@ int SSLWrap<Base>::SelectALPNCallback(SSL* s,\n unsigned alpn_protos_len = Buffer::Length(alpn_buffer);\n int status = SSL_select_next_proto(const_cast<unsigned char**>(out), outlen,\n alpn_protos, alp... | 2017-09-23T16:44:24 |
golang/go | 1c0d1f4e975eb864e36b18e63bcb8cdedbdfc51b | 6af088bfc66c13143c9ef46b4cf0805df77a8fbe | cmd/compile: optimize regalloc for phi value
When allocating registers for phi value, only the primary predecessor is considered.
Taking into account the allocation status of other predecessors can help reduce
unnecessary copy or spill operations. Many such cases can be found in the standard
library, such as runtime.w... | [
{
"path": "src/cmd/compile/internal/ssa/regalloc.go",
"patch": "@@ -1012,8 +1012,8 @@ func (s *regAllocState) regalloc(f *Func) {\n \t\t\t// Copy phi ops into new schedule.\n \t\t\tb.Values = append(b.Values, phis...)\n \n-\t\t\t// Third pass - pick registers for phis whose inputs\n-\t\t\t// were not in a r... | 2020-10-19T03:20:24 |
ollama/ollama | af8a8a6b59080f2309e7761e995d7b83bfb8c941 | 461ad25015cc9b760a49bdb8e8fd26c87c2a359c | fix: trim quotes on OLLAMA_ORIGINS | [
{
"path": "server/routes.go",
"patch": "@@ -1013,16 +1013,14 @@ func allowedHostsMiddleware(addr net.Addr) gin.HandlerFunc {\n }\n \n func (s *Server) GenerateRoutes() http.Handler {\n-\tvar origins []string\n-\tif o := os.Getenv(\"OLLAMA_ORIGINS\"); o != \"\" {\n-\t\torigins = strings.Split(o, \",\")\n-\t}... | 2024-03-27T22:24:28 |
huggingface/transformers | 3b9174f248a9c16e05a6701f1c5bdc28fc19f995 | ea50b64bea95b80cb3e8fcceaeb8304b90b5900c | interpolation added for TVP. (#30863)
* Update TVP model to interpolate pre-trained image pad prompter encodings
* feat: Add 2D positional embeddings interpolation in TvpVisualInputEmbedding
* added required comments
* Update TVP model to interpolate pre-trained image pad prompter encodings
* feat: Add 2D ... | [
{
"path": "src/transformers/models/tvp/modeling_tvp.py",
"patch": "@@ -193,34 +193,81 @@ def __init__(self, config):\n self.token_type_embeddings = nn.Embedding(1, config.hidden_size)\n self.layer_norm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)\n self.dropout = nn.... | 2024-06-07T17:44:16 |
nodejs/node | d9b9229d98afb4b3f01e73c619c5fede0f87138f | e433afa65245d2aa2eff84da1c2f69db83460cbe | test: fix test-https-agent-session-eviction for 1.1
This test is testing the workaround for an OpenSSL 1.0.x bug, which was
fixed in 1.1.0. With the bug fixed, the test expectations need to change
slightly.
PR-URL: https://github.com/nodejs/node/pull/16130
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: ... | [
{
"path": "src/node_constants.cc",
"patch": "@@ -759,6 +759,10 @@ void DefineSignalConstants(Local<Object> target) {\n }\n \n void DefineOpenSSLConstants(Local<Object> target) {\n+#ifdef OPENSSL_VERSION_NUMBER\n+ NODE_DEFINE_CONSTANT(target, OPENSSL_VERSION_NUMBER);\n+#endif\n+\n #ifdef SSL_OP_ALL\n ... | 2017-09-23T08:07:58 |
golang/go | 6af088bfc66c13143c9ef46b4cf0805df77a8fbe | 612a363bef9ae29d190f6daa2a5a1623f78c874b | encoding/json: add "json: " prefix to SyntaxError messages
The other named errors - UnmarshalTypeError, etc - in this package do
the same, so we should prepend the package prefix to error messages
for consistency.
Add a note to the release docs in case this is interpreted as
a breaking change.
Fixes #36221.
Change-... | [
{
"path": "doc/go1.16.html",
"patch": "@@ -216,6 +216,14 @@ <h3 id=\"crypto/x509\"><a href=\"/pkg/crypto/x509\">crypto/x509</a></h3>\n of a malformed certificate.\n </p>\n \n+<h3 id=\"encoding/json\"><a href=\"/pkg/encoding/json\">encoding/json</a></h3>\n+\n+<p><!-- CL 263619 -->\n+ The error message for... | 2020-10-19T20:29:40 |
ggml-org/llama.cpp | 9b24886f78ce278b34186b47ed71a435f00d8d0d | 62b8143ad217fc7d3404119a9440be4a18ce31e0 | Fix compile bug (#20203)
* Fix compile bug
* Update common/chat-auto-parser-helpers.cpp
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
---------
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com> | [
{
"path": "common/chat-auto-parser-helpers.cpp",
"patch": "@@ -162,7 +162,7 @@ diff_split calculate_diff_split(const std::string & left, const std::string & ri\n right_fully_consumed = true;\n }\n \n- auto eat_segment = [](std::string & str, segment & seg) -> std::string { return str.append(s... | 2026-03-08T16:15:49 |
rust-lang/rust | c867b8f11dbf0be397ba9b5c898fde95f8d5ea6c | 30f168ef811aec63124eac677e14699baa9395bd | Construct MIR error body for global_asm correctly | [
{
"path": "compiler/rustc_mir_build/src/builder/mod.rs",
"patch": "@@ -612,7 +612,8 @@ fn construct_error(tcx: TyCtxt<'_>, def_id: LocalDefId, guar: ErrorGuaranteed) -\n | DefKind::AssocConst\n | DefKind::AnonConst\n | DefKind::InlineConst\n- | DefKind::Static { .. } => (vec![... | 2025-02-23T22:54:36 |
huggingface/transformers | 065729a692c3cf80e2e8eb995aa1122f6b1b10bc | 3a10058201de322f3b66c85476cca2f49905650f | Remove ConversationalPipeline and Conversation object (#31165)
* Remove ConversationalPipeline and Conversation object, as they have been deprecated for some time and are due for removal
* Update not-doctested.txt
* Fix JA and ZH docs
* Fix JA and ZH docs some more
* Fix JA and ZH docs some more | [
{
"path": "docs/source/en/main_classes/pipelines.md",
"patch": "@@ -386,14 +386,6 @@ Pipelines available for computer vision tasks include the following.\n \n Pipelines available for natural language processing tasks include the following.\n \n-### ConversationalPipeline\n-\n-[[autodoc]] Conversation\n-\n-[... | 2024-06-07T16:50:18 |
golang/go | 612a363bef9ae29d190f6daa2a5a1623f78c874b | b04eb73a6891e81e2e611ee93b3aa8c4769d5967 | cmd/compile/internal/gc: fix comments
Change-Id: Id7b0ead39e961a16a85da3e308db10dd4f9b55c3
GitHub-Last-Rev: e640c4a61ade361ac17b7eb95d0ce8913d0b4d6f
GitHub-Pull-Request: golang/go#42120
Reviewed-on: https://go-review.googlesource.com/c/go/+/264080
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Cuong Manh Le <cuong... | [
{
"path": "src/cmd/compile/internal/gc/dcl.go",
"patch": "@@ -283,7 +283,7 @@ func oldname(s *types.Sym) *Node {\n \t\t\tc.Name.Defn = n\n \n \t\t\t// Link into list of active closure variables.\n-\t\t\t// Popped from list in func closurebody.\n+\t\t\t// Popped from list in func funcLit.\n \t\t\tc.Name.Para... | 2020-10-21T16:18:34 |
ggml-org/llama.cpp | 62b8143ad217fc7d3404119a9440be4a18ce31e0 | d088d5b74f1d63b9a345d1515ab9e3bb3bc81a10 | Fix structured outputs (#20223)
* Fix structured outputs
* Update common/chat-auto-parser-generator.cpp
Co-authored-by: Aldehir Rojas <hello@alde.dev>
---------
Co-authored-by: Aldehir Rojas <hello@alde.dev> | [
{
"path": "common/chat-auto-parser-generator.cpp",
"patch": "@@ -1,6 +1,7 @@\n #include \"chat-auto-parser.h\"\n #include \"chat-peg-parser.h\"\n #include \"chat.h\"\n+#include \"common.h\"\n #include \"json-schema-to-grammar.h\"\n #include \"nlohmann/json.hpp\"\n \n@@ -51,13 +52,15 @@ common_chat_params pe... | 2026-03-08T16:14:43 |
huggingface/transformers | ff689f57aa111261e6c2a506a42479d99674b123 | 8bcf9c8dd4419b6b0216fa68c2a5b2b156cd64de | Extend save_pretrained to offloaded models (#27412)
* added hidden subset
* debugged hidden subset contrastive search
* added contrastive search compression
* debugged compressed contrastive search
* memory reduction for contrastive search
* debugged mem red
* added low memory option feature
* deb... | [
{
"path": "src/transformers/modeling_utils.py",
"patch": "@@ -119,6 +119,10 @@\n set_module_tensor_to_device,\n )\n \n+ accelerate_version = version.parse(importlib.metadata.version(\"accelerate\"))\n+ if accelerate_version >= version.parse(\"0.31\"):\n+ from accelerate.utils.modeli... | 2024-06-07T11:50:35 |
ggml-org/llama.cpp | d088d5b74f1d63b9a345d1515ab9e3bb3bc81a10 | cd18a50ea573c081e7fa5604c85e4d571fd6ae4f | ggml-vulkan: Add ELU op support (#20183)
* ggml-Vulkan: add ELU support
* ggml-Vulkan: remove extra spaces and variables
* ggml-Vulkan: fix format issue
* ggml-Vulkan: fix format issue
* fix whitespace issue
* Update Vulkan.csv and ops.md | [
{
"path": "docs/ops.md",
"patch": "@@ -37,11 +37,11 @@ Legend:\n | CROSS_ENTROPY_LOSS | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ |\n | CROSS_ENTROPY_LOSS_BACK | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ |\n | CUMSUM | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ... | 2026-03-08T11:38:17 |
golang/go | 214136b7412d56151d5443741feb0ed873facf2e | bcc333348769efed7c38acfa013e5475c53e8f5f | cmd/go/internal/fsys: use a root other than "." in Walk tests
Fixes #42115
Change-Id: Icf4c9eac5ed3295acbc8377c7a06f82c6bddc747
Reviewed-on: https://go-review.googlesource.com/c/go/+/264177
Trust: Bryan C. Mills <bcmills@google.com>
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.c... | [
{
"path": "src/cmd/go/internal/fsys/fsys_test.go",
"patch": "@@ -487,6 +487,11 @@ contents don't matter for this test\n }\n \n func TestWalk(t *testing.T) {\n+\t// The root of the walk must be a name with an actual basename, not just \".\".\n+\t// Walk uses Lstat to obtain the name of the root, and Lstat on... | 2020-10-21T14:07:02 |
ollama/ollama | b8c2be6142068b540406d2f9c84b1576b4394b0a | e0319bd78d834675a06fb2e9e2673efe29dfa722 | Use Rocky Linux Vault to get GCC 10.2 installed
This should hopefully only be a temporary workaround until Rocky 8
picks up GCC 10.4 which fixes the NVCC bug | [
{
"path": "scripts/rh_linux_deps.sh",
"patch": "@@ -20,7 +20,19 @@ if grep -i \"centos\" /etc/system-release >/dev/null; then\n fi\n dnf install -y devtoolset-10-gcc devtoolset-10-gcc-c++\n elif grep -i \"rocky\" /etc/system-release >/dev/null; then\n- dnf install -y git gcc-toolset-10-gcc gcc-to... | 2024-03-26T02:18:50 |
huggingface/transformers | 8bcf9c8dd4419b6b0216fa68c2a5b2b156cd64de | f868cf731a9bf60853a7a543dbe8472173aa59cc | Fix jetmoe model (#31279)
* Fix jetmoe model
* Remove skip-tests | [
{
"path": "src/transformers/models/jetmoe/modeling_jetmoe.py",
"patch": "@@ -1404,18 +1404,14 @@ def prepare_inputs_for_generation(\n \n past_length = 0\n if past_key_values is not None:\n- if isinstance(past_key_values, Cache):\n- past_length = cache_position[0] if... | 2024-06-07T09:51:41 |
ggml-org/llama.cpp | cd18a50ea573c081e7fa5604c85e4d571fd6ae4f | a976ff081b4657b67f48295bbefc030d9d899b17 | vulkan: Fix data races in coopmat1 mul_mat(_id) (#20084)
* vulkan: Fix data races in coopmat1 mul_mat(_id)
Add barriers between coopmat store and regular loads. We sort of got away with
this because it was the same subgroup accessing the values, but it's still a
race and may not work.
* switch to subgroup control ba... | [
{
"path": "ggml/src/ggml-vulkan/vulkan-shaders/mul_mm.comp",
"patch": "@@ -377,6 +377,7 @@ void main() {\n [[unroll]] for (uint cm_col = 0; cm_col < cms_per_col; cm_col++) {\n coopMatStore(sums[cm_col * cms_per_row + cm_row], coopmat_stage, warp_i * TM * TN, TM, gl_CooperativeMatrixLayou... | 2026-03-08T11:33:48 |
golang/go | bcc333348769efed7c38acfa013e5475c53e8f5f | 15ead857dbc638b9d83a7686acf0dc746fc45918 | cmd/go: ignore GOFLAGS values without name in go env/bug
This happens with 'go env' and 'go bug'.
If GOFLAGS variable is set to something like '=value',
running `go env` panics with this error message:
goroutine 1 [running]:
cmd/go/internal/base.SetFromGOFLAGS(0xd96838)
cmd/go/internal/base/goflags.g... | [
{
"path": "src/cmd/go/internal/base/goflags.go",
"patch": "@@ -92,7 +92,11 @@ func SetFromGOFLAGS(flags *flag.FlagSet) {\n \t}\n \tfor _, goflag := range goflags {\n \t\tname, value, hasValue := goflag, \"\", false\n-\t\tif i := strings.Index(goflag, \"=\"); i >= 0 {\n+\t\t// Ignore invalid flags like '=' o... | 2020-10-16T15:34:15 |
nodejs/node | 00872a851f9be8fccaed42b8f22f8e2c8f68012b | f72975ffa29b7f3a8d4210bf6b34ca625e6b29f2 | test: test with a larger RSA key
OpenSSL 1.1.0 rejects RSA keys smaller than 1024 bits by default. Fix
the tests to use larger ones. This test only cares that the PEM blob be
missing a trailing newline. Certificate adapted from
test/fixtures/cert.pem.
PR-URL: https://github.com/nodejs/node/pull/16130
Reviewed-By: Ben... | [
{
"path": "test/parallel/test-tls-cert-regression.js",
"patch": "@@ -27,29 +27,43 @@ if (!common.hasCrypto)\n \n const tls = require('tls');\n \n-\n const cert =\n `-----BEGIN CERTIFICATE-----\n-MIIBfjCCASgCCQDmmNjAojbDQjANBgkqhkiG9w0BAQUFADBFMQswCQYDVQQGEwJB\n-VTETMBEGA1UECBMKU29tZS1TdGF0ZTEhMB8GA1UEChMYSW... | 2017-09-23T07:03:26 |
huggingface/transformers | f868cf731a9bf60853a7a543dbe8472173aa59cc | bdf36dcd48106a4a0278ed7f3cc26cd65ab7b066 | Fixed Wav2Vec2ProcessorWithLM decoding error (#31188)
* fix: wav2vec2_with_lm decoding error
Fixed an error where some language models could
not be loaded due to a decoding error, since it
was impossible to select the 'unigram_encoding'
value.
* fix: unexpected keyword argument
Fixed unexpected keyword arg... | [
{
"path": "src/transformers/models/wav2vec2_with_lm/processing_wav2vec2_with_lm.py",
"patch": "@@ -152,7 +152,8 @@ def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):\n feature_extractor, tokenizer = super()._get_arguments_from_pretrained(pretrained_model_name_or_path, **kwargs)\n \n ... | 2024-06-07T09:50:07 |
ollama/ollama | f38b705dc7b18817020c0ae19cbb58d915e67324 | 560be5e0b60741600140fa0918ee125f0b7f9c75 | Fix ROCm link in `development.md` | [
{
"path": "docs/development.md",
"patch": "@@ -69,7 +69,7 @@ go build .\n \n _Your operating system distribution may already have packages for AMD ROCm and CLBlast. Distro packages are often preferable, but instructions are distro-specific. Please consult distro-specific docs for dependencies if available!_... | 2024-03-25T20:32:44 |
nodejs/node | ffb8f031a4559f89b6967e3a1eb0cbdb24657bb0 | 5fe81c8aff03261f6443580dbc08f608013718c6 | test: update test expectations for OpenSSL 1.1.0
Some errors in the two versions are different. The test-tls-no-sslv3 one
because OpenSSL 1.1.x finally does version negotiation properly. 1.0.x's
logic was somewhat weird and resulted in very inconsistent errors for
SSLv3 in particular.
Also the function codes are capi... | [
{
"path": "test/parallel/test-crypto.js",
"patch": "@@ -238,7 +238,7 @@ assert.throws(function() {\n // Throws crypto error, so there is an opensslErrorStack property.\n // The openSSL stack should have content.\n if ((err instanceof Error) &&\n- /asn1 encoding routines:ASN1_CHECK_TLEN:wrong tag/... | 2017-09-17T22:48:25 |
golang/go | 15ead857dbc638b9d83a7686acf0dc746fc45918 | 54c0237346adfc2cac7bbebba80d652227ab6ea5 | cmd/compiler,cmd/go,sync: add internal {LoadAcq,StoreRel}64 on ppc64
Add an internal atomic intrinsic for load with acquire semantics
(extending LoadAcq to 64b) and add LoadAcquintptr for internal
use within the sync package. For other arches, this remaps to the
appropriate atomic.Load{,64} intrinsic which should not... | [
{
"path": "src/cmd/compile/internal/gc/ssa.go",
"patch": "@@ -3389,6 +3389,13 @@ func init() {\n \t\t\treturn s.newValue1(ssa.OpSelect0, types.Types[TUINT32], v)\n \t\t},\n \t\tsys.PPC64, sys.S390X)\n+\taddF(\"runtime/internal/atomic\", \"LoadAcq64\",\n+\t\tfunc(s *state, n *Node, args []*ssa.Value) *ssa.Va... | 2020-09-09T22:24:23 |
huggingface/transformers | bdf36dcd48106a4a0278ed7f3cc26cd65ab7b066 | a3d351c00f815aa71b70829951c6137efe1101da | Enable HF pretrained backbones (#31145)
* Enable load HF or tim backbone checkpoints
* Fix up
* Fix test - pass in proper out_indices
* Update docs
* Fix tvp tests
* Fix doc examples
* Fix doc examples
* Try to resolve DPT backbone param init
* Don't conditionally set to None
* Add condition... | [
{
"path": "docs/source/en/create_a_model.md",
"patch": "@@ -327,31 +327,21 @@ For example, to load a [ResNet](../model_doc/resnet) backbone into a [MaskFormer\n Set `use_pretrained_backbone=True` to load pretrained ResNet weights for the backbone.\n \n ```py\n-from transformers import MaskFormerConfig, Mask... | 2024-06-06T21:02:38 |
ggml-org/llama.cpp | a976ff081b4657b67f48295bbefc030d9d899b17 | a95047979a1671be970398a7c8073159ac71013e | llama: end-to-end tests (#19802)
* tests: add end-to-end tests per model architecture
* fixup for rebase
* fix use-after-free in llama-model-loader.cpp
* fix CI
* fix WebGPU
* fix CI
* disable CI for macOS-latest-cmake-arm64
* use expert_weights_scale only if != 0.0f
* comments | [
{
"path": ".github/workflows/build.yml",
"patch": "@@ -93,7 +93,7 @@ jobs:\n id: cmake_test\n run: |\n cd build\n- ctest -L main --verbose --timeout 900\n+ ctest -L main -E \"test-llama-archs\" --verbose --timeout 900\n \n macOS-latest-cmake-x64:\n runs-on: ... | 2026-03-08T11:30:21 |
ollama/ollama | 7b6cbc10ec9765b0013c163e868c5307d9d27009 | acfa2b94220882ffe8246a56e35a6cd1db38e6a2 | Integration tests conditionally pull
If images aren't present, pull them.
Also fixes the expected responses | [
{
"path": "integration/basic_test.go",
"patch": "@@ -12,7 +12,7 @@ import (\n )\n \n func TestOrcaMiniBlueSky(t *testing.T) {\n-\tctx, cancel := context.WithTimeout(context.Background(), 1*time.Minute)\n+\tctx, cancel := context.WithTimeout(context.Background(), 2*time.Minute)\n \tdefer cancel()\n \t// Set ... | 2024-03-24T23:22:38 |
nodejs/node | afee2ed490a458709514d686978d911fe6b7810c | 87a066a108db1e5e93a5a4a705b2f4bcbb1dcfb2 | crypto: use RSA and DH accessors
Parts of this were cherry-picked from PR #8491. Note that this only
works with OpenSSL 1.0.2 or 1.1.0g or later. 1.1.0g is, as of writing,
not yet released, but the fix is on the branch. See
https://github.com/openssl/openssl/pull/4384.
PR-URL: https://github.com/nodejs/node/pull/1613... | [
{
"path": "src/node_crypto.cc",
"patch": "@@ -107,6 +107,77 @@ using v8::Value;\n \n \n #if OPENSSL_VERSION_NUMBER < 0x10100000L\n+static void RSA_get0_key(const RSA* r, const BIGNUM** n, const BIGNUM** e,\n+ const BIGNUM** d) {\n+ if (n != nullptr) {\n+ *n = r->n;\n+ }\n+ if (... | 2017-09-17T23:19:50 |
golang/go | 54c0237346adfc2cac7bbebba80d652227ab6ea5 | 9848e93cb76f068b97e8ea342c33cc09b553790f | misc/wasm: improve error message if javascript polyfills are required
wasm_exec.js expects that either "require" is available or that the
globals "crypto", "TextEncoder" and "TextDecoder" are already defined.
Report a better error message if this is not the case, suggesting the
use of a polyfill.
Updates #41482
Chan... | [
{
"path": "misc/wasm/wasm_exec.js",
"patch": "@@ -102,14 +102,17 @@\n \t\t}\n \t}\n \n-\tif (!global.crypto) {\n+\tif (!global.crypto && global.require) {\n \t\tconst nodeCrypto = require(\"crypto\");\n \t\tglobal.crypto = {\n \t\t\tgetRandomValues(b) {\n \t\t\t\tnodeCrypto.randomFillSync(b);\n \t\t\t},\n \... | 2020-10-11T08:23:45 |
huggingface/transformers | 3b4d3d09fd8cee2b4cc2fdd7c12ea51ca147c6cc | b6c9f47fd6f911450024c52e382e544e5d04387a | Fix SwinLayer / DonutSwinLayer / ClapAudioLayer attention mask device (#31295)
Fix DonutSwinLayer attention mask device | [
{
"path": "src/transformers/models/clap/modeling_clap.py",
"patch": "@@ -593,10 +593,10 @@ def set_shift_and_window_size(self, input_resolution):\n self.shift_size = 0\n self.window_size = min(input_resolution)\n \n- def get_attn_mask(self, height, width, dtype):\n+ def get_att... | 2024-06-06T20:52:14 |
ggml-org/llama.cpp | 649f06481e363fa02a53b89af9659645730c367b | 746368716193bbbfa75abb6da81284bfce99539e | quants : Add memsets and other fixes for IQ quants (#19861)
* Add memsets and other fixes for IQ quants
* Make memset unconditional, change Laux back to L
* Move another memset | [
{
"path": "ggml/src/ggml-quants.c",
"patch": "@@ -3104,6 +3104,11 @@ static void quantize_row_iq2_xxs_impl(const float * GGML_RESTRICT x, void * GGML\n }\n float scale = make_qp_quants(32, kMaxQ+1, xval, (uint8_t*)L, weight);\n float eff_max = scale*kMaxQ;\n+ i... | 2026-03-06T21:06:56 |
ollama/ollama | 6ce37e4d969b65e9af8e7b3bd3989412daf5bee5 | 703684a82a150ba59b300ca9c1ec678bd9b4ab27 | llm,readline: use errors.Is instead of simple == check (#3161)
This fixes some brittle, simple equality checks to use errors.Is. Since
go1.13, errors.Is is the idiomatic way to check for errors.
Co-authored-by: Jeffrey Morgan <jmorganca@gmail.com> | [
{
"path": "llm/gguf.go",
"patch": "@@ -3,6 +3,7 @@ package llm\n import (\n \t\"bytes\"\n \t\"encoding/binary\"\n+\t\"errors\"\n \t\"fmt\"\n \t\"io\"\n \t\"log/slog\"\n@@ -540,7 +541,7 @@ func (llm *GGUFModel) Encode(f *os.File) error {\n \t\t\tb, err := io.ReadFull(dataFile, data)\n \t\t\tremaining -= uint... | 2024-03-15T14:14:12 |
huggingface/transformers | f9296249a3d8fabf9d77d33dbc1105037ead6164 | 4c821025238735d94c9e2ff09b81ae6f79c1db15 | Pipeline VQA: Add support for list of images and questions as pipeline input (#31217)
* Add list check for image and question
* Handle passing two lists and update docstring
* Add tests
* Add support for dataset
* Add test for dataset as input
* fixup
* fix unprotected import
* fix unprotected imp... | [
{
"path": "src/transformers/pipelines/visual_question_answering.py",
"patch": "@@ -1,4 +1,4 @@\n-from typing import Union\n+from typing import List, Union\n \n from ..utils import add_end_docstrings, is_torch_available, is_vision_available, logging\n from .base import Pipeline, build_pipeline_init_args\n@@ ... | 2024-06-06T13:50:45 |
golang/go | 9848e93cb76f068b97e8ea342c33cc09b553790f | 491fe1b959bf6a0657cba2e3db831ffd4bdf7674 | misc/wasm: make sure sp is unsigned
An i32 passed from WebAssembly to JavaScript is always read as a signed
integer. Use the bitshift operator to turn it into an unsigned integer.
Fixes #40923
Change-Id: Ia91ed2145dd2fc3071e2fc22b86ebfcb3c1e9f4f
Reviewed-on: https://go-review.googlesource.com/c/go/+/261358
Trust: Ri... | [
{
"path": "misc/wasm/wasm_exec.js",
"patch": "@@ -255,6 +255,7 @@\n \n \t\t\t\t\t// func wasmExit(code int32)\n \t\t\t\t\t\"runtime.wasmExit\": (sp) => {\n+\t\t\t\t\t\tsp >>>= 0;\n \t\t\t\t\t\tconst code = this.mem.getInt32(sp + 8, true);\n \t\t\t\t\t\tthis.exited = true;\n \t\t\t\t\t\tdelete this._inst;\n@... | 2020-10-11T11:33:53 |
nodejs/node | 0f842f500eb7e117428e2227482f864ae72f58f0 | 5d94712b859339d48b3eb2d3bdd00d04d66f6160 | crypto: remove unnecessary SSLerr calls
These are OpenSSL-internal APIs that are no longer accessible in 1.1.0
and weren't necessary. OpenSSL will push its own errors and, if it
doesn't, the calling code would handle it anyway.
PR-URL: https://github.com/nodejs/node/pull/16130
Reviewed-By: Ben Noordhuis <info@bnoordh... | [
{
"path": "src/node_crypto.cc",
"patch": "@@ -661,7 +661,6 @@ int SSL_CTX_use_certificate_chain(SSL_CTX* ctx,\n x = PEM_read_bio_X509_AUX(in, nullptr, NoPasswordCallback, nullptr);\n \n if (x == nullptr) {\n- SSLerr(SSL_F_SSL_CTX_USE_CERTIFICATE_CHAIN_FILE, ERR_R_PEM_LIB);\n return 0;\n }\n \n@... | 2017-09-16T07:16:17 |
huggingface/transformers | 99895ae5e265ea67be77d9bc51544b093105424c | 5ba8ac54f5674dc0d6010bbe48c7e72389052c69 | fix accelerate tests for roberta xl (#31288)
* fix accelerate tests for roberta xl
* style | [
{
"path": "src/transformers/models/xlm_roberta_xl/modeling_xlm_roberta_xl.py",
"patch": "@@ -572,7 +572,7 @@ class XLMRobertaXLPreTrainedModel(PreTrainedModel):\n \n config_class = XLMRobertaXLConfig\n base_model_prefix = \"roberta\"\n- _no_split_modules = [\"XLMRobertaXLEmbeddings\", \"XLMRobert... | 2024-06-06T13:44:35 |
ollama/ollama | 703684a82a150ba59b300ca9c1ec678bd9b4ab27 | 6459377ae081021a0ec2b5b408c6ac4503de01b2 | server: replace blob prefix separator from ':' to '-' (#3146)
This fixes issues with blob file names that contain ':' characters to be rejected by file systems that do not support them. | [
{
"path": "server/fixblobs.go",
"patch": "@@ -0,0 +1,26 @@\n+package server\n+\n+import (\n+\t\"os\"\n+\t\"path/filepath\"\n+\t\"strings\"\n+)\n+\n+// fixBlobs walks the provided dir and replaces (\":\") to (\"-\") in the file\n+// prefix. (e.g. sha256:1234 -> sha256-1234)\n+func fixBlobs(dir string) error ... | 2024-03-15T03:18:06 |
nodejs/node | e0b1394cb00a6e4f5a5d4daca486f5bb64836e5a | b9a772116b457bee59612bcaf489d95c61397a6c | doc: fix typos in N-API
PR-URL: https://github.com/nodejs/node/pull/16911
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewe... | [
{
"path": "doc/api/n-api.md",
"patch": "@@ -1788,7 +1788,7 @@ This API returns the C int32 primitive equivalent\n of the given JavaScript Number. If the number exceeds the range of the\n 32 bit integer, then the result is truncated to the equivalent of the\n bottom 32 bits. This can result in a large positi... | 2017-11-10T03:44:18 |
ggml-org/llama.cpp | 566059a26b0ce8faec4ea053605719d399c64cc5 | 34df42f7bef5a711b2b40f5d2b6b78254def99c3 | Autoparser - complete refactoring of parser architecture (#18675)
* Autoparser - full single commit squish
* Final pre-merge changes: minor fixes, Kimi 2.5 model parser | [
{
"path": "common/CMakeLists.txt",
"patch": "@@ -47,10 +47,10 @@ add_library(${TARGET} STATIC\n arg.cpp\n arg.h\n base64.hpp\n- chat-parser.cpp\n- chat-parser.h\n- chat-parser-xml-toolcall.h\n- chat-parser-xml-toolcall.cpp\n+ chat-auto-parser-generator.cpp\n+ chat-auto-parser-h... | 2026-03-06T20:01:00 |
huggingface/transformers | 5ba8ac54f5674dc0d6010bbe48c7e72389052c69 | 14ff5dd962c1bd0a4e3adaac347ba396d8df5add | Fix _save_tpu: use _maybe_convert_to_cpu instead of to cpu. (#31264)
* Fix _save_tpu: use _maybe_convert_to_cpu instead of to cpu.
* fix lint | [
{
"path": "src/transformers/trainer.py",
"patch": "@@ -3407,8 +3407,6 @@ def _save_tpu(self, output_dir: Optional[str] = None):\n logger.info(f\"Saving model checkpoint to {output_dir}\")\n model = self.model\n xm.mark_step()\n- if self.args.save_safetensors:\n- mod... | 2024-06-06T13:42:55 |
ggml-org/llama.cpp | 34df42f7bef5a711b2b40f5d2b6b78254def99c3 | e68f2fb894d890eeead6acf0cc3341478312f1fd | hexagon: add f32 ssm_conv op (#20122)
* hexagon: add ssm_conv op
* hexagon: hvx kernel is functional
* hexagon: improvements to ssm-conv hvx kernel
* hexagon: added dma to ssm-conv hvx kernel
* hexagon: ssm-conv dynamically compute gather scratchpad
* hex-ssm-conv: add local context and fix various issues (spad i... | [
{
"path": "ggml/src/ggml-hexagon/ggml-hexagon.cpp",
"patch": "@@ -2152,6 +2152,44 @@ static bool ggml_hexagon_supported_rope(const struct ggml_hexagon_session * sess\n return true;\n }\n \n+static bool ggml_hexagon_supported_ssm_conv(const struct ggml_hexagon_session * sess, const struct ggml_tensor * o... | 2026-03-06T17:59:26 |
ollama/ollama | 8546dd3d72867be55c7439ecc58bc0790a0f556b | 87100be5e0a12a9b058f331b50509e041c266738 | .github: fix model and feature request yml (#3155) | [
{
"path": ".github/ISSUE_TEMPLATE/10_model_request.yml",
"patch": "@@ -1,6 +1,5 @@\n name: Model request\n description: Request a new model for the library\n-title: \"\"\n labels: [mr]\n body:\n - type: markdown",
"additions": 0,
"deletions": 1,
"language": "YAML"
},
{
"path": ".github... | 2024-03-14T22:26:06 |
nodejs/node | b340c2d4dd9d774dbed35ca1c33678ff30d46192 | 3cacd34dc2c5da4c01c8fd09c34a816357358c8b | build: remove cctest extension
cctest has `so.59` extension when building node shared library in linux.
The appending is defined in node.gypi and the cctest target in node.gyp
includes node.gypi. Moving the appending from node.gypi to node target
in node.gyp fixes the issue.
Signed-off-by: Yihong Wang <yh.wang@ibm.co... | [
{
"path": "node.gyp",
"patch": "@@ -315,6 +315,11 @@\n 'NODE_OPENSSL_SYSTEM_CERT_PATH=\"<(openssl_system_ca_path)\"',\n ],\n },\n+ 'conditions': [\n+ [ 'node_shared==\"true\" and node_module_version!=\"\" and OS!=\"win\"', {\n+ 'product_extension': '<(shlib_suffi... | 2017-11-02T04:19:16 |
huggingface/transformers | 9e9679c022790d74a7588e844a7598e3590aea7e | 9ef93fccade1f2b6349c2eec2188d0ea54329f84 | fix: `str` should be used not `int` when setting env variables (#31272) | [
{
"path": "src/transformers/trainer_utils.py",
"patch": "@@ -72,8 +72,8 @@ def enable_full_determinism(seed: int, warn_only: bool = False):\n os.environ[\"CUDA_LAUNCH_BLOCKING\"] = \"1\"\n os.environ[\"CUBLAS_WORKSPACE_CONFIG\"] = \":16:8\"\n # The environment variable required to en... | 2024-06-06T11:41:31 |
ggml-org/llama.cpp | d48e876467734ab9c4292340a32b52c92660111c | ba2ff79e43bad81fd19ecc8324431c93499e459e | ggml-cuda: add mem check for fusion (#19916)
* ggml-cuda: add mem check for fusion
* Replace NaNs with -FLT_MAX
* fix typo
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
---------
Co-authored-by: Johannes Gäßler <johannesg@5d6.de> | [
{
"path": "ggml/src/ggml-cuda/ggml-cuda.cu",
"patch": "@@ -3412,6 +3412,69 @@ static bool ggml_cuda_can_fuse(const struct ggml_cgraph * cgraph,\n return false;\n }\n \n+// returns whether the write (out) nodes overwrite the read nodes in operation\n+static bool ggml_cuda_check_fusion_memo... | 2026-03-06T16:05:43 |
golang/go | 1d07306a447ae50d3eae5eed6ecc4f30a2528c96 | befc62a2c47859d4cb63ee1f9b6e0b7f95a4c50a | [dev.typeparams] cmd/compile/internal/types2: adjust tests, enable Testdata tests
Types2 uses a different test runner and has fewer/better
errors in some cases (error messages match the compiler).
Adjust the tests and enable them.
Change-Id: I74877f54a81a3918a80774452cef5bcaad8a98e6
Reviewed-on: https://go-review.goo... | [
{
"path": "src/cmd/compile/internal/types2/check_test.go",
"patch": "@@ -221,8 +221,7 @@ func TestCheck(t *testing.T) {\n \tcheckFiles(t, strings.Split(*testFiles, \" \"), 0, testing.Verbose())\n }\n \n-// TODO(gri) Enable once we have added the testdata tests.\n-// func TestTestdata(t *testing.T) { DefPre... | 2020-10-20T04:38:31 |
nodejs/node | 3cacd34dc2c5da4c01c8fd09c34a816357358c8b | 766cd1f59d5adb03953656391e63b4186c9aff10 | src: add process.ppid
Fixes: https://github.com/nodejs/node/issues/14957
PR-URL: https://github.com/nodejs/node/pull/16839
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com> | [
{
"path": "doc/api/process.md",
"patch": "@@ -1416,6 +1416,19 @@ system platform on which the Node.js process is running. For instance\n console.log(`This platform is ${process.platform}`);\n ```\n \n+## process.ppid\n+<!-- YAML\n+added: REPLACEME\n+-->\n+\n+* {integer}\n+\n+The `process.ppid` property retu... | 2017-10-30T20:14:15 |
huggingface/transformers | 5fabd1e83bfa6f0df144b3aee2987ccb70aec973 | 7729b7747872eabe461a8d2d4ce1068ae0e716a8 | Generation: fix handling of special tokens (#31254)
* fix special tokens in generatioon
* fix test
* add warning
* fix the check
* warn once
* fix | [
{
"path": "src/transformers/generation/utils.py",
"patch": "@@ -1436,23 +1436,6 @@ def _get_cache(self, cache_implementation: str, max_batch_size: int, max_cache_l\n self._cache.reset()\n return self._cache\n \n- def _get_decoder_start_token_id(\n- self, decoder_start_token_id:... | 2024-06-06T10:21:32 |
ggml-org/llama.cpp | c6980ff29ddc8e59c9c002dcaeec14182d893ed7 | 1e38a7a6fa115de0a2731cb67ce554b7df5e8e2c | ggml-cpu: Fix gcc 15 ICE on ppc64le (#20083) (#20130)
This patch addresses an Internal Compiler Error (Segmentation fault)
observed with gcc 15 by replacing the intrinsic + cast by doing
a cat on the data first and then calling the intrinsic. This bypasses the
buggy compiler path while maintaining identical instructio... | [
{
"path": "ggml/src/ggml-cpu/llamafile/sgemm.cpp",
"patch": "@@ -2497,7 +2497,7 @@ class tinyBLAS_Q0_PPC {\n for (int r = 0; r < 8; r++) {\n const block_q4_0 * current_blk = rows_base[r] + blk;\n vector float v_scale = vec_extract_fp32_from_shorth(vec_... | 2026-03-06T15:22:39 |
golang/go | cfb087c68324e0af4217ffe75a962ca8383d12c5 | f121e0eddd64aa0084b916e94c9a540edf859539 | cmd/go: fix TestScript/test_cache_inputs
Small fix for a new failure in CL 263142
For #41190
Change-Id: I733bc1998d87b505b52d429916c45afe2968a709
Reviewed-on: https://go-review.googlesource.com/c/go/+/264057
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Jay Conrod <jay... | [
{
"path": "src/cmd/go/testdata/script/test_cache_inputs.txt",
"patch": "@@ -137,7 +137,7 @@ exit 0\n package testcache\n \n import (\n-\t\"io/ioutil\"\n+\t\"io\"\n \t\"os\"\n \t\"testing\"\n )",
"additions": 1,
"deletions": 1,
"language": "Plain Text"
}
] | 2020-10-20T22:18:21 |
nodejs/node | ccab7d1dac9d625a4d94b0093b212bc7a8c461c6 | 7af469a605a90260951817bc1e8e370fb28b897e | deps: cherry-pick cc55747 from V8 upstream
This fixes the flaky message/console test on our CI.
Original commit message:
[test/message] Allow numbers to have more than one leading digit.
The {NUMBER} regexp only allowed one, leading to occasional test
failures such as:
https://build.chromium.org/p/client.v8/... | [
{
"path": "common.gypi",
"patch": "@@ -27,7 +27,7 @@\n \n # Reset this number to 0 on major V8 upgrades.\n # Increment by one for each non-official patch applied to deps/v8.\n- 'v8_embedder_string': '-node.9',\n+ 'v8_embedder_string': '-node.10',\n \n # Enable disassembler for `--print-cod... | 2017-11-08T19:09:54 |
huggingface/transformers | 7729b7747872eabe461a8d2d4ce1068ae0e716a8 | f5c0fa9f6fe0eea2ad69bb1b03aff04824aa4870 | Make mamba use cache (#31116)
* make mamba use cache
* uss cache naming as in mamba
* fix musicgen | [
{
"path": "src/transformers/generation/utils.py",
"patch": "@@ -627,18 +627,22 @@ def _expand_dict_for_generation(dict_to_expand):\n \n def _extract_past_from_model_output(self, outputs: ModelOutput, standardize_cache_format: bool = False):\n past_key_values = None\n+ cache_name = \"past_... | 2024-06-06T08:37:29 |
ggml-org/llama.cpp | 2850bc6a1324fb810aecaf236dc5955a1c142a15 | 17a42589467165be7114b79797e794716b30ace3 | ggml-cpu: fix data race for debug asserts (#20148) | [
{
"path": "ggml/src/ggml-cpu/ops.cpp",
"patch": "@@ -2129,12 +2129,12 @@ static void ggml_compute_forward_gelu_f32(\n \n #ifndef NDEBUG\n for (int k = 0; k < nc; k++) {\n- const float x = ((float *) ((char *) dst->data + i1*( dst->nb[1])))[k];\n+ const float x = ((float *) ((ch... | 2026-03-06T08:12:49 |
golang/go | f121e0eddd64aa0084b916e94c9a540edf859539 | 3ca3ca51d70aac890fb023786db1fcbbeea139f1 | cmd/compile: fix nodedump output for types of nodes
The Dbg dumping of complex types was broken, because (I think) of a
recent change to handle recursive types correctly. Before this fix,
the Dump output of a closure node (where the last thing on the line is
the type of the node) was:
. . CLOSURE l(8) esc(h) tc(1... | [
{
"path": "src/cmd/compile/internal/gc/fmt.go",
"patch": "@@ -792,6 +792,13 @@ func tconv2(b *bytes.Buffer, t *types.Type, flag FmtFlag, mode fmtMode, visited\n \t\treturn\n \t}\n \n+\tif mode == FDbg {\n+\t\tb.WriteString(t.Etype.String())\n+\t\tb.WriteByte('-')\n+\t\ttconv2(b, t, flag, FErr, visited)\n+\t... | 2020-10-20T22:03:33 |
huggingface/transformers | f5c0fa9f6fe0eea2ad69bb1b03aff04824aa4870 | 9b85e405abcfe30e0c7b72b18805aa44d75f118d | fix loading special_tokens_map_file (#31012) | [
{
"path": "src/transformers/tokenization_utils_base.py",
"patch": "@@ -2285,11 +2285,14 @@ def _from_pretrained(\n # We keep this new value and ignore the one stored in the special_tokens_map_file\n continue\n if isinstance(valu... | 2024-06-06T07:15:27 |
ollama/ollama | 291c663865fe9765aa4c541e617948a54d400590 | da20786e3e22a9f5a7223081ccfec200cfbd5255 | fix: clip memory leak | [
{
"path": "llm/patches/05-fix-clip-free.diff",
"patch": "@@ -0,0 +1,45 @@\n+From 9192432daf90b1bfec75577434a99b4ea70d54c8 Mon Sep 17 00:00:00 2001\n+From: Michael Yang <mxyng@pm.me>\n+Date: Thu, 14 Mar 2024 12:09:50 -0700\n+Subject: [PATCH] fix clip free\n+\n+---\n+ examples/llava/clip.cpp | 4 ++++\n+ ex... | 2024-03-14T19:45:46 |
ggml-org/llama.cpp | 17a42589467165be7114b79797e794716b30ace3 | f7db3f37895ad3a98405afdb84b42ed462542b12 | kv-cache : fix M-RoPE checkpoints (#20132) | [
{
"path": "src/llama-batch.cpp",
"patch": "@@ -394,11 +394,13 @@ llama_ubatch llama_batch_allocr::ubatch_reserve(uint32_t n_seq_tokens, uint32_t\n clear();\n split_reset();\n \n+ const int64_t n_pos_all = (int64_t) n_tokens*n_pos_per_embd;\n+\n auto udata = std::make_shared<llama_ubatch::data... | 2026-03-06T06:46:51 |
nodejs/node | 7af469a605a90260951817bc1e8e370fb28b897e | cad1d1ff457afb93034b995111ef1b8cc1f08d82 | doc: fix a typo in the documentation
PR-URL: https://github.com/nodejs/node/pull/16909
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com> | [
{
"path": "doc/api/deprecations.md",
"patch": "@@ -122,7 +122,7 @@ to the `constants` property exposed by the relevant module. For instance,\n Type: End-of-life\n \n Use of the [`crypto.pbkdf2()`][] API without specifying a digest was deprecated\n-in Node.js 6.0 because the method defaulted to using the non... | 2017-11-10T04:42:58 |
golang/go | b7a2d413a3f710f14accedf185c93bfb63d24fd0 | 80182d45b5d2ff86da7b6587a2a09d8924dd0a95 | testing: print cpu type as label for benchmarks
Supports 386 and amd64 architectures on all operating systems.
Example output:
$ go test -bench=.*
goos: darwin
goarch: amd64
pkg: strconv
cpu: Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz
BenchmarkAtof64Decimal-4 24431032 46.8 ns/op
...
As the displayed CP... | [
{
"path": "src/go/build/deps_test.go",
"patch": "@@ -467,6 +467,10 @@ var depsRules = `\n \t< net/rpc\n \t< net/rpc/jsonrpc;\n \n+\t# System Information\n+\tinternal/cpu, sync\n+\t< internal/sysinfo;\n+\n \t# Test-only\n \tlog\n \t< testing/iotest\n@@ -475,7 +479,7 @@ var depsRules = `\n \tFMT, flag, math/r... | 2020-10-20T07:56:14 |
huggingface/transformers | 9b85e405abcfe30e0c7b72b18805aa44d75f118d | 8177aa0e1ac58595c1af35b493a11ae681d8c55c | [`SwitchTransformer`] Significant performance improvement on MoE blocks (#31173)
* SwitchTransformer MoE layer performance improvement
* make fixup
* comments about shapes
* make fixup | [
{
"path": "src/transformers/models/switch_transformers/modeling_switch_transformers.py",
"patch": "@@ -294,9 +294,17 @@ def forward(self, hidden_states):\n # can be unchanged from one layer to another. That is why the hidden states are cloned before updating only the seleced ones.\n \n next_... | 2024-06-06T07:10:12 |
ggml-org/llama.cpp | 2b10b62677af99fcd07b69aa05561e499bb070dc | a0ed91a442ea6b013bd42ebc3887a81792eaefa1 | hexagon: add fp16 support for binary ops: add,sub,mul,div (#20139)
* hexagon: add fp16 support for binary ops: add,sub,mul,div
* hexagon: fix test-backend-ops failures for fp16 binary ops on older arches (<v79)
* hexagon: decide on n_threads (aka n_jobs) early to avoid overallocating scratchpad
* snapdragon: fix re... | [
{
"path": "README.md",
"patch": "@@ -287,7 +287,7 @@ Instructions for adding support for new models: [HOWTO-add-model.md](docs/develo\n | [IBM zDNN](docs/backend/zDNN.md) | IBM Z & LinuxONE |\n | [WebGPU [In Progress]](docs/build.md#webgpu) | All |\n | [RPC](https://github.com/ggml-org/llama.cpp/tree/master... | 2026-03-06T02:29:13 |
golang/go | c9c64886ef041b096d7f93c4e7d2ef5faf87ad43 | e2c420591cbbd684594a111fa5cdeb40c68964a5 | cmd/internal/obj: reject too large symbols
We never supported symbol larger than 2GB (issue #9862), so the
object file uses 32-bit for symbol sizes. Check and reject too
large symbol before truncating its size.
Fixes #42054.
Change-Id: I0d1d585ebdba9556f2fd3a97043bd4296d5cc9e4
Reviewed-on: https://go-review.googleso... | [
{
"path": "src/cmd/internal/obj/objfile.go",
"patch": "@@ -261,6 +261,10 @@ func (w *writer) StringTable() {\n \t}\n }\n \n+// cutoff is the maximum data section size permitted by the linker\n+// (see issue #9862).\n+const cutoff = int64(2e9) // 2 GB (or so; looks better in errors than 2^31)\n+\n func (w *w... | 2020-10-19T14:40:24 |
ggml-org/llama.cpp | a0ed91a442ea6b013bd42ebc3887a81792eaefa1 | 2cd20b72ed3565ac6935911ca0d9b5d73ae70d0d | models : kda chunk size = 16 (#19827)
* models : add llm_build_delta_net_base
* cont : keep qwen35 and qwen35moe graphs intact
* cont : add comments [no ci]
* add kimi linear to delta-net-base
* removed unnecessary ggml_cont from g_exp_t
* removed ggml_cont from g_diff_exp_t. moved ggml_cont for o to kimi-linear.... | [
{
"path": "src/models/delta-net-base.cpp",
"patch": "@@ -1,7 +1,5 @@\n #include \"models.h\"\n \n-#define CHUNK_SIZE 64\n-\n // utility to get one slice from the third dimension\n // input dim: [x, y, c, b]\n // output dim: [x, y, 1, b]\n@@ -57,7 +55,7 @@ std::pair<ggml_tensor *, ggml_tensor *> llm_build_d... | 2026-03-05T15:01:23 |
ollama/ollama | 82b0c7c27e50c17b6fa5418caf56d32da28a9049 | 34d00f90b1cf9db0c59c953b3128f58d25073b36 | Fix iGPU detection for linux
This fixes a few bugs in the new sysfs discovery logic. iGPUs are now
correctly identified by their <1G VRAM reported. the sysfs IDs are off
by one compared to what HIP wants due to the CPU being reported
in amdgpu, but HIP only cares about GPUs. | [
{
"path": "gpu/amd_common.go",
"patch": "@@ -40,19 +40,17 @@ func amdSetVisibleDevices(ids []int, skip map[int]interface{}) {\n \t// TODO - does sort order matter?\n \tdevices := []string{}\n \tfor i := range ids {\n-\t\tslog.Debug(fmt.Sprintf(\"i=%d\", i))\n \t\tif _, skipped := skip[i]; skipped {\n-\t\t\t... | 2024-03-12T23:57:19 |
ggml-org/llama.cpp | 872646b30c342c5fa98efec002f94fdb4f1f1ea5 | b5ed0e058c98645c1cab752d012fdc4be22bceef | model : update Qwen3.5 model type detection (#20126)
* model : fix Qwen3.5 model type detection
* Update src/llama-model.cpp
whoops, my bad
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
---------
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com> | [
{
"path": "src/llama-model.cpp",
"patch": "@@ -61,6 +61,7 @@ const char * llm_type_name(llm_type type) {\n case LLM_TYPE_0_3B: return \"0.3B\";\n case LLM_TYPE_0_5B: return \"0.5B\";\n case LLM_TYPE_0_6B: return \"0.6B\";\n+ case LLM_TYPE_0_8B: ... | 2026-03-05T11:47:14 |
golang/go | e2c420591cbbd684594a111fa5cdeb40c68964a5 | 1b09d430678d4a6f73b2443463d11f75851aba8a | cmd/go/internal/modload: remove printStackInDie functionality
Previously, when running cmd/go tests, if the module root directory is
requested when modules are explicitly disabled, we printed a stack
trace in addition to the error message that's normally printed. The
stack trace isn't that useful, and it makes the act... | [
{
"path": "src/cmd/go/internal/modload/build.go",
"patch": "@@ -13,7 +13,6 @@ import (\n \t\"internal/goroot\"\n \t\"os\"\n \t\"path/filepath\"\n-\t\"runtime/debug\"\n \t\"strings\"\n \n \t\"cmd/go/internal/base\"\n@@ -312,9 +311,6 @@ func mustFindModule(target, path string) module.Version {\n \t\treturn Ta... | 2020-10-19T18:02:14 |
ollama/ollama | 53c107e20e183e3fd59f046275a6a95d6f6d4ddc | 51578d857319a1e807e746dc7ab1b1f30a444958 | chore: fix typo (#3073)
Signed-off-by: racerole <jiangyifeng@outlook.com> | [
{
"path": "llm/ext_server/ext_server.cpp",
"patch": "@@ -115,7 +115,7 @@ void llama_server_init(ext_server_params *sparams, ext_server_resp_t *err) {\n llama_numa_init(params.numa);\n \n if (!llama->load_model(params)) { \n- // an error occured that was not thrown\n+ // an error occurred that wa... | 2024-03-12T18:09:22 |
nodejs/node | 44d3e17985befbd45457d5ad7f0a0387849e1b2f | 3b3ceafaf922e1d79950595eaa501aa412913820 | deps: ICU 60 bump
- Update to released ICU 60.1, including:
- CLDR 32 (many new languages and data improvements)
- Unicode 10 (8,518 new characters, including four new scripts,
7,494 new Han characters, and 56 new emoji characters)
- UTF-8 malformed bytes now handled according to W3C/WHATWG spec
Fixes: https:... | [
{
"path": "LICENSE",
"patch": "@@ -230,7 +230,7 @@ The externally maintained libraries used by Node.js are:\n # ---------COPYING.libtabe ---- BEGIN--------------------\n #\n # /*\n- # * Copyrighy (c) 1999 TaBE Project.\n+ # * Copyright (c) 1999 TaBE Project.\n # * Copyrig... | 2017-09-21T22:31:38 |
ggml-org/llama.cpp | 92f7da00b49ad814b95832dd6610a825bbdd3033 | 7a99dc85e2d26a9b9c540cad887322eae8924f03 | chore : correct typos [no ci] (#20041)
* fix(docs): correct typos found during code review
Non-functional changes only:
- Fixed minor spelling mistakes in comments
- Corrected typos in user-facing strings
- No variables, logic, or functional code was modified.
Signed-off-by: Marcel Petrick <mail@marcelpetrick.it>
*... | [
{
"path": "CONTRIBUTING.md",
"patch": "@@ -159,7 +159,7 @@ Maintainers reserve the right to decline review or close pull requests for any r\n \n # Code maintenance\n \n-- Existing code should have designated collaborators and/or maintainers specified in the [CODEOWNERS](CODEOWNERS) file reponsible for:\n+- ... | 2026-03-05T07:50:21 |
rust-lang/rust | 5d259224bd69425b44b4c83adc310cbf72b15855 | 329b8a312dc5fbe8a310a30b8ce452c7609e202c | Manual, post-`clippy --fix` cleanups | [
{
"path": "src/librustdoc/clean/mod.rs",
"patch": "@@ -2539,7 +2539,7 @@ fn clean_generic_args<'tcx>(\n ) -> GenericArgs {\n // FIXME(return_type_notation): Fix RTN parens rendering\n if let Some((inputs, output)) = generic_args.paren_sugar_inputs_output() {\n- let inputs = inputs.iter().map(... | 2025-03-01T19:12:21 |
ollama/ollama | 51578d857319a1e807e746dc7ab1b1f30a444958 | b5fcd9d3aa6b05fe09f0a604bde9314d5d8d27ea | fix gpu_info_cuda.c compile warning (#3077) | [
{
"path": "gpu/gpu_info_cuda.c",
"patch": "@@ -155,8 +155,8 @@ void cuda_check_vram(cuda_handle_t h, mem_info_t *resp) {\n }\n }\n \n- LOG(h.verbose, \"[%d] CUDA totalMem %ld\\n\", i, memInfo.total);\n- LOG(h.verbose, \"[%d] CUDA usedMem %ld\\n\", i, memInfo.used);\n+ LOG(h.verbose, \"[%d... | 2024-03-12T18:08:40 |
huggingface/transformers | bd5091df8db7cea1a9f94f797fc11487f840ade1 | d6276f0fc5d32d741374fd28615feb6be594cbfe | Reduce by 2 the memory requirement in `generate()` 🔥🔥🔥 (#30536)
* Fix contrastive_search for new cache structure, and improve performance by removing inneficient torch.stack(torch.split(x, top_k, dim=0))
* Fix _contrastive_search for non-standard cache using ellipsis slicing
* Fix all outputs.logits memory le... | [
{
"path": "src/transformers/cache_utils.py",
"patch": "@@ -377,22 +377,75 @@ def get_max_length(self) -> Optional[int]:\n return None\n \n def to_legacy_cache(self) -> Tuple[Tuple[torch.Tensor], Tuple[torch.Tensor]]:\n- \"\"\"Converts the `DynamicCache` instance into the its equivalent in... | 2024-06-05T15:05:01 |
nodejs/node | f823d381e700ec152575ebd2cdc4420def690833 | e5238ed030a06fa94e4574eff504777e173c9351 | src: fix UB in InternalModuleReadFile()
`&vec[0]` is undefined behavior when `vec.size() == 0`.
It is mostly academic because package.json files are not usually empty
and because with most STL implementations it decays to something that
is legal C++ as long as the result is not dereferenced, but better safe
than sorr... | [
{
"path": "src/node_file.cc",
"patch": "@@ -515,12 +515,17 @@ static void InternalModuleReadFile(const FunctionCallbackInfo<Value>& args) {\n start = 3; // Skip UTF-8 BOM.\n }\n \n- Local<String> chars_string =\n- String::NewFromUtf8(env->isolate(),\n- &chars[start],\n-... | 2017-11-07T20:28:13 |
ggml-org/llama.cpp | 7a99dc85e2d26a9b9c540cad887322eae8924f03 | 69fd345335a5d637f72ccb765c9b5fe0b848da48 | hexagon: Flash Attention optimizations (dma, mpyacc, multi-row) and MatMul updates (#20118)
* ggml-hexagon: enhance hvx_dot_f16_f16_aa_rx4 for improved performance by expanding vector handling and optimizing accumulation
# Conflicts:
# ggml/src/ggml-hexagon/htp/flash-attn-ops.c
* ggml-hexagon: optimize hvx_dot_f16_f... | [
{
"path": "ggml/src/ggml-hexagon/htp/flash-attn-ops.c",
"patch": "@@ -10,13 +10,24 @@\n \n #include \"hex-dma.h\"\n #include \"hvx-utils.h\"\n+#include \"hvx-dump.h\"\n \n #define GGML_COMMON_DECL_C\n #include \"ggml-common.h\"\n #include \"htp-ctx.h\"\n #include \"htp-msg.h\"\n #include \"htp-ops.h\"\n \n+... | 2026-03-05T05:55:29 |
rust-lang/rust | ccfbfe22922946249088bc9d574f7dc22c8f6c90 | 30f168ef811aec63124eac677e14699baa9395bd | `x clippy src/librustdoc --fix` | [
{
"path": "src/librustdoc/clean/mod.rs",
"patch": "@@ -2539,7 +2539,7 @@ fn clean_generic_args<'tcx>(\n ) -> GenericArgs {\n // FIXME(return_type_notation): Fix RTN parens rendering\n if let Some((inputs, output)) = generic_args.paren_sugar_inputs_output() {\n- let inputs = inputs.iter().map(... | 2025-03-06T16:06:41 |
ollama/ollama | b80661e8c78e115ed9b41391c87fdb7f1a7f69ec | 6d3adfbea21699db7770eb608264d5e16b8663ee | relay load model errors to the client (#3065) | [
{
"path": "llm/dyn_ext_server.go",
"patch": "@@ -149,7 +149,7 @@ func newDynExtServer(library, model string, adapters, projectors []string, opts\n \n \tslog.Info(\"Initializing llama server\")\n \tslog.Debug(fmt.Sprintf(\"server params: %+v\", sparams))\n-\tinitResp := newExtServerResp(128)\n+\tinitResp := ... | 2024-03-11T20:48:27 |
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.