Remove env variable BIGDL_LLM_XMX_DISABLED in documentation (#12445)

* fix: remove BIGDL_LLM_XMX_DISABLED in mddocs * fix: remove set SYCL_CACHE_PERSISTENT=1 in example * fix: remove BIGDL_LLM_XMX_DISABLED in workflows * fix: merge igpu and A-series Graphics * fix: remove set BIGDL_LLM_XMX_DISABLED=1 in example * fix: remove BIGDL_LLM_XMX_DISABLED in workflows * fix: merge igpu and A-series Graphics * fix: textual adjustment * fix: textual adjustment * fix: textual adjustment
intel-analytics · Nov 27, 2024 · acd77d9 · acd77d9
1 parent f8c2bb2
commit acd77d9
Show file tree

Hide file tree

Showing 122 changed files with 119 additions and 1,443 deletions.
diff --git a/.github/workflows/llm_performance_tests.yml b/.github/workflows/llm_performance_tests.yml
@@ -805,7 +805,6 @@ jobs:
           call conda activate igpu-perf
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -841,7 +840,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -875,7 +873,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -910,7 +907,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -947,7 +943,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1003,7 +998,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1039,7 +1033,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1073,7 +1066,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1108,7 +1100,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1145,7 +1136,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1200,7 +1190,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1236,7 +1225,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1270,7 +1258,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1305,7 +1292,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1342,7 +1328,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1397,7 +1382,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1435,7 +1419,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1474,7 +1457,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1522,7 +1504,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1572,7 +1553,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1642,7 +1622,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1680,7 +1659,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1717,7 +1695,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1756,7 +1733,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1814,7 +1790,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1852,7 +1827,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1888,7 +1862,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1925,7 +1898,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -1964,7 +1936,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -2021,7 +1992,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -2059,7 +2029,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -2095,7 +2064,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -2132,7 +2100,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1
@@ -2171,7 +2138,6 @@ jobs:
 
           if "${{ matrix.platform }}"=="perf-mtl" (
             set SYCL_CACHE_PERSISTENT=1
-            set BIGDL_LLM_XMX_DISABLED=1
           )
           if "${{ matrix.platform }}"=="perf-lnl" (
             set SYCL_CACHE_PERSISTENT=1

diff --git a/docs/mddocs/DockerGuides/docker_pytorch_inference_gpu.md b/docs/mddocs/DockerGuides/docker_pytorch_inference_gpu.md
@@ -97,10 +97,6 @@ root@arda-arc12:/# sycl-ls
 > # Reduce memory accesses by fusing SDP ops.
 > # Recommended for use on Intel Data Center GPU Max Series.
 > export ENABLE_SDP_FUSION=1
-> 
-> # Disable XMX computation.
-> # Recommended for use on integrated GPUs.
-> export BIGDL_LLM_XMX_DISABLED=1
 > ```
 
 

diff --git a/docs/mddocs/Overview/install_gpu.md b/docs/mddocs/Overview/install_gpu.md
@@ -150,13 +150,7 @@ call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat"
 
 Please also set the following environment variable if you would like to run LLMs on: -->
 
-- For **Intel iGPU**:
-   ```cmd
-   set SYCL_CACHE_PERSISTENT=1
-   set BIGDL_LLM_XMX_DISABLED=1
-   ```
-
-- For **Intel Arc™ A-Series Graphics**:
+- For **Intel iGPU** and **Intel Arc™ A-Series Graphics**:
    ```cmd
    set SYCL_CACHE_PERSISTENT=1
    ```
@@ -596,7 +590,6 @@ To use GPU acceleration on Linux, several environment variables are required or
       source /opt/intel/oneapi/setvars.sh
 
       export SYCL_CACHE_PERSISTENT=1
-      export BIGDL_LLM_XMX_DISABLED=1
       ```
 
 > [!NOTE]

diff --git a/docs/mddocs/Quickstart/benchmark_quickstart.md b/docs/mddocs/Quickstart/benchmark_quickstart.md
@@ -80,34 +80,19 @@ Some parameters in the yaml file that you can configure:
 
 Please refer to [here](../Overview/install_gpu.md#runtime-configuration) to configure oneAPI environment variables. Choose corresponding commands base on your device.
 
-- For **Intel iGPU**:
+- For **Intel iGPU** and **Intel Arc™ A-Series Graphics**:
 
   ```bash
   set SYCL_CACHE_PERSISTENT=1
-  set BIGDL_LLM_XMX_DISABLED=1
   
   python run.py
   ```
 
-- For **Intel Arc™ A300-Series or Pro A60**:
-
-  ```bash
-  set SYCL_CACHE_PERSISTENT=1
-  python run.py
-  ```
-
-- For **Other Intel dGPU Series**:
-
-  ```bash
-  # e.g. Arc™ A770
-  python run.py
-  ```
-
 ## Run on Linux
 
 Please choose corresponding commands base on your device.
 
-- For **Intel Arc™ A-Series and Intel Data Center GPU Flex**:
+- For **Intel Arc™ A-Series** and **Intel Data Center GPU Flex**:
 
   For Intel Arc™ A-Series Graphics and Intel Data Center GPU Flex Series, we recommend:
 

diff --git a/docs/mddocs/Quickstart/install_linux_gpu.md b/docs/mddocs/Quickstart/install_linux_gpu.md
@@ -422,7 +422,6 @@ To use GPU acceleration on Linux, several environment variables are required or
   source /opt/intel/oneapi/setvars.sh
 
   export SYCL_CACHE_PERSISTENT=1
-  export BIGDL_LLM_XMX_DISABLED=1
   ```
 
 > [!NOTE]

diff --git a/docs/mddocs/Quickstart/install_linux_gpu.zh-CN.md b/docs/mddocs/Quickstart/install_linux_gpu.zh-CN.md
@@ -390,7 +390,6 @@ conda activate llm
   source /opt/intel/oneapi/setvars.sh
 
   export SYCL_CACHE_PERSISTENT=1
-  export BIGDL_LLM_XMX_DISABLED=1
   ```
 
 > [!NOTE]

diff --git a/docs/mddocs/Quickstart/install_windows_gpu.md b/docs/mddocs/Quickstart/install_windows_gpu.md
@@ -109,19 +109,12 @@ You can verify if `ipex-llm` is successfully installed following below steps.
 
 - Set the following environment variables according to your device:
 
-  - For **Intel iGPU**:
+  - For **Intel iGPU** and **Intel Arc™ A770**:
 
     ```cmd
     set SYCL_CACHE_PERSISTENT=1
-    set BIGDL_LLM_XMX_DISABLED=1
     ```
-
-  - For **Intel Arc™ A770**:
-
-    ```cmd
-    set SYCL_CACHE_PERSISTENT=1
-    ```
-  
+    
 > [!TIP]
 > For other Intel dGPU Series, please refer to [this guide](../Overview/install_gpu.md#runtime-configuration) for more details regarding runtime configuration.
 

diff --git a/docs/mddocs/Quickstart/install_windows_gpu.zh-CN.md b/docs/mddocs/Quickstart/install_windows_gpu.zh-CN.md
@@ -108,14 +108,7 @@ conda activate llm
 
 - 根据你的设备，设置以下环境参数：
 
-  - **Intel iGPU**:
-
-    ```cmd
-    set SYCL_CACHE_PERSISTENT=1
-    set BIGDL_LLM_XMX_DISABLED=1
-    ```
-
-  - **Intel Arc™ A770**:
+  - **Intel iGPU** and **Intel Arc™ A770**:
 
     ```cmd
     set SYCL_CACHE_PERSISTENT=1