-
TLDR: CUDA memory API is slow, and Halide allocates and frees all intermediate buffers on the fly _within_ the pipeline. Hence, buffer reuse is not in my control. Envisioned elegant solution: have a b…
-
When creating the parser context, the memory functions set when creating the xmpp-context is not forwarded to libexpat. This causes libexpat to use regular malloc/realloc/free instead, which is a issu…
-
To find memory leaks in zig programs, the standard library allocators should inform valgrind about allocations/frees.
This has a dependency on making valgrind client requests possible from zig code…
-
We've been getting intermittent crashes in a custom build of Electron 29.3.1 on openSUSE Tumbleweed x64 with the following stack trace:
```log
#0 0x00005583d8b3e4f7 in non-virtual thunk to electron…
-
vcpkg.json
```json
{
"$schema": "https://raw.githubusercontent.com/microsoft/vcpkg/master/scripts/vcpkg.schema.json",
"name": "jsdfgksdjf",
"version": "0.0.1",
"builtin-baseline"…
-
### Description
`patch_inplace` has several instances of `get`, which breaks the build if you're using a custom string.
### Reproduction steps
- Instantiate a nlohmann::basic_json with a cust…
-
This is a tracking issue for the `try_reserve` part of the RFC "fallible collection allocation" (rust-lang/rfcs#2116).
**Steps:**
- [x] Implement the RFC #48648
- [x] Add `HashSet::try_reserv…
-
## 🐛 Bug
Using `torch.repeat` leads to runtime error:
```
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1717594747.366986 228 cuda_exec…
-
Struggling to install and build test.py on Windows. I search all issues.
Using Python 3.12.
I did install `conan==1.62.0`.
The linux instructions have `sudo apt-get install shedskin` which t…
-
For some reason when I manually build, it's giving a weird bug.
```
Boost.Python.ArgumentError: Python argument types in
simple_object_detector.save(simple_object_detector, str)
did not match C…