summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJiří Sejkora <Sejseloid@gmail.com>2024-04-10 00:23:02 +0200
committerGitHub <noreply@github.com>2024-04-10 00:23:02 +0200
commitba5e134e073ec6837078c874aba44a702944a676 (patch)
tree0b263ccf08476d9b58e7a657c0b283ee9376209b
parent1b67731e184e27a465b8c5476061294a4af668ea (diff)
readme: fix typo in amdgpu target name (#6573)
-rw-r--r--README.md2
1 files changed, 1 insertions, 1 deletions
diff --git a/README.md b/README.md
index a4897fc3..b6066238 100644
--- a/README.md
+++ b/README.md
@@ -513,7 +513,7 @@ Building the program with BLAS support may lead to some performance improvements
- Using `make` (example for target gfx1030, build with 16 CPU threads):
```bash
- make -j16 LLAMA_HIPBLAS=1 LLAMA_HIP_UMA=1 AMDGPU_TARGETS=gxf1030
+ make -j16 LLAMA_HIPBLAS=1 LLAMA_HIP_UMA=1 AMDGPU_TARGETS=gfx1030
```
- Using `CMake` for Windows (using x64 Native Tools Command Prompt for VS, and assuming a gfx1100-compatible AMD GPU):