-
### Issue type
Build/Install
### Have you reproduced the bug with TensorFlow Nightly?
Yes
### Source
source
### TensorFlow version
v2.14.0-4246-g996b6d313c1 2.14.0
### Custom code
No
### OS …
-
### 🐛 Describe the bug
I cannot build pytorch from source using Fedora 40's supplied version of ROCM.
```
Building PyTorch for GPU arch: gfx1030
CMake Warning at cmake/public/LoadHIP.cmake:36 (f…
-
After conv numerics issue #498 got fixed with this [fix](https://github.com/openxla/iree/pull/16640)
We see an error in the output of the first SDPA Op. when running UNet
The numerical error for …
-
This code (seen in pytorch) compiles with stdlibc++, but fails with libc++:
```c++
// hipcc -c test.cpp -o test.o
#include
__device__ void test1( void* dest, const void* src, std::size_t coun…
-
### Problem Description
Pytorch fails to compile locally with aotriton, and throws the following error:
```
make -j 6 -f Makefile.shim HIPCC=hipcc AR=/usr/bin/ar EXTRA_COMPILER_OPTIONS=-I/opt/rocm…
-
- How will it be built?
- How will it be maintained?
https://github.com/mgaitan/sphinxcontrib-mermaid
https://devblogs.microsoft.com/cppblog/vcpkg-integration-with-the-github-dependency-graph/
-
When compiling with `-Wpendantic`, `g++` 13.1 emits the following warnings when simply including `hip/hip_runtime.h`:
```
g++ --std=c++17 playground.cpp -Wall -Wextra -Wpedantic -o playground -I/o…
-
Running `clEnqueueSVMMemcpy(queue, CL_TRUE, dst, src, size, 0, NULL, NULL);` with dst allocated with `clSVMAlloc` and src allocated by the system (e.g. posix_memalign) triggers a segmentation fault:
…
-
I have not been able to get `llamafile --gpu amd` working with an `AMD Radeon RX 6900 XT` on `linux`. The relevant line of the log seems to be:
```
llamafile: /usr/src/debug/hip-runtime-amd/clr-ro…
-
On Ubuntu 22.04, kernel 6.7.9, ROCm 6.1.0 (RC), Radeon Pro VII.
In brief: when creating a second command-queue (that is not even used at all) one thread starts eating 100% CPU, i.e. doing busy-wait…
preda updated
1 month ago