Skip to content

Commit b36301c

Browse files
docs: update Hygon DCU support to all models and fix Manufacturer typo
Rebase onto upstream master with a clean single commit, removing unrelated drift (LinkedIn footer revert and blog formatting changes). Signed-off-by: kinglong08 <ise_longlong@163.com> Co-authored-by: Cursor <cursoragent@cursor.com>
1 parent a3e36d4 commit b36301c

24 files changed

Lines changed: 100 additions & 100 deletions

File tree

docs/contributor/roadmap.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,7 @@ sidebar_label: Roadmap
1111
| GPU | NVIDIA | All | Yes | Yes | Yes |
1212
| MLU | Cambricon | 370, 590 | Yes | Yes | No |
1313
| GCU | Enflame | S60 | Yes | Yes | No |
14-
| DCU | Hygon | Z100, Z100L | Yes | Yes | No |
14+
| DCU | Hygon | All | Yes | Yes | No |
1515
| NPU | Huawei Ascend | 310P, 910B, 910B3 | Yes | Yes | No |
1616
| GPU | Iluvatar | All | Yes | Yes | No |
1717
| DPU | Teco | Checking | In progress | In progress | No |

docs/faq/faq.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@ title: FAQ
88
| --- | --- | --- | --- |
99
| NVIDIA | Almost all mainstream consumer and data center GPUs | Core 1%, Memory 1M | Supported. Multi-GPU can still be split and shared using virtualization. |
1010
| Huawei Ascend | 910A, 910B2, 910B3, 310P | Minimum granularity depends on the card type template. Refer to the [official templates](https://www.hiascend.com/document/detail/zh/mindx-dl/50rc1/AVI/cpaug/cpaug_0005.html). | Supported, but splitting is not supported when `npu > 1`. The entire card is exclusively allocated. |
11-
| Hygon | Z100, Z100L, K100-AI | Core 1%, Memory 1M | Supported, but splitting is not supported when `dcu > 1`. The entire card is exclusively allocated. |
11+
| Hygon | All | Core 1%, Memory 1M | Supported, but splitting is not supported when `dcu > 1`. The entire card is exclusively allocated. |
1212
| Cambricon | 370, 590 | Core 1%, Memory 256M | Supported, but splitting is not supported when `mlu > 1`. The entire card is exclusively allocated. |
1313
| Iluvatar | All | Core 1%, Memory 256M | Supported, but splitting is not supported when `gpu > 1`. The entire card is exclusively allocated. |
1414
| Moore Threads | MTT S4000 | Core 1 core group, Memory 512M | Supported, but splitting is not supported when `gpu > 1`. The entire card is exclusively allocated. |

docs/userguide/device-supported.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ The table below lists the devices supported by HAMi:
99
| ---- | ------------- | ----------------------- | --------------- | ------------- | ----------------- |
1010
| GPU | NVIDIA | All | Yes | Yes | Yes |
1111
| MLU | Cambricon | 370, 590 | Yes | Yes | No |
12-
| DCU | Hygon | Z100, Z100L | Yes | Yes | No |
12+
| DCU | Hygon | All | Yes | Yes | No |
1313
| NPU | Huawei Ascend | 910B, 910B3, 910C, 310P | Yes | Yes | No |
1414
| GPU | Iluvatar | All | Yes | Yes | No |
1515
| GPU | Mthreads | MTT S4000 | Yes | Yes | No |

i18n/zh/docusaurus-plugin-content-docs/current/faq/faq.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@ title: 常见问题
88
| --- | --- | --- | --- |
99
| NVIDIA | 几乎所有主流消费级和数据中心 GPU | 核心 1%,显存 1M | 支持。多 GPU 仍可通过虚拟化进行拆分和共享。 |
1010
| 昇腾 | 910A、910B2、910B3、310P | 最小粒度取决于卡类型模板。参考[官方模板](https://www.hiascend.com/document/detail/zh/mindx-dl/50rc1/AVI/cpaug/cpaug_0005.html)| 支持,但当 `npu > 1` 时不支持拆分,整卡独占。 |
11-
| 海光 | Z100、Z100L、K100-AI | 核心 1%,显存 1M | 支持,但当 `dcu > 1` 时不支持拆分,整卡独占。 |
11+
| 海光 | 全部 | 核心 1%,显存 1M | 支持,但当 `dcu > 1` 时不支持拆分,整卡独占。 |
1212
| 寒武纪 | 370、590 | 核心 1%,显存 256M | 支持,但当 `mlu > 1` 时不支持拆分,整卡独占。 |
1313
| 天数智芯 | 全部 | 核心 1%,显存 256M | 支持,但当 `gpu > 1` 时不支持拆分,整卡独占。 |
1414
| 摩尔线程 | MTT S4000 | 核心为 1 个核心组,显存 512M | 支持,但当 `gpu > 1` 时不支持拆分,整卡独占。 |

i18n/zh/docusaurus-plugin-content-docs/current/userguide/device-supported.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@ HAMi 支持的设备如下表所示:
1010
| -------- | ------------------------- | --------------------------- | -------- | -------- | -------- |
1111
| GPU | 英伟达(NVIDIA) | 全系列 ||||
1212
| MLU | 寒武纪(Cambricon) | 370、590 ||||
13-
| DCU | 海光(Hygon) | Z100、Z100L ||||
13+
| DCU | 海光(Hygon) | 全系列 ||||
1414
| NPU | 华为昇腾(Huawei Ascend) | 910B、910B3、910C、310P ||||
1515
| GPU | 天数智芯(Iluvatar) | 全部 ||||
1616
| GPU | 摩尔线程(Mthreads) | MTT S4000 ||||

i18n/zh/docusaurus-plugin-content-docs/version-v1.3.0/userguide/device-supported.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -4,11 +4,11 @@ title: Device supported by HAMi
44

55
The view of device supported by HAMi is shown in this table below:
66

7-
| Production | manufactor | Type |MemoryIsolation | CoreIsolation | MultiCard support |
7+
| Production | Manufacturer | Type |MemoryIsolation | CoreIsolation | MultiCard support |
88
|-------------|------------|-------------|-----------|---------------|-------------------|
99
| GPU | NVIDIA | All | Yes | Yes | Yes |
1010
| MLU | Cambricon | 370, 590 | Yes | Yes | No |
11-
| DCU | Hygon | Z100, Z100L | Yes | Yes | No |
11+
| DCU | Hygon | All | Yes | Yes | No |
1212
| Huawei Ascend | Huawei | 910B, 910B3, 310P | Yes | Yes | No |
1313
| GPU | iluvatar | All | Yes | Yes | No |
1414
| GPU | Mthreads | MTT S4000 | Yes | Yes | No |

i18n/zh/docusaurus-plugin-content-docs/version-v2.5.0/userguide/device-supported.md

Lines changed: 8 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -5,12 +5,12 @@ translated: true
55

66
HAMi 支持的设备如下表所示:
77

8-
| 类型 | 制造商 | 型号 | 内存隔离 | 核心隔离 | 多卡支持 |
9-
|-----|-------|-----|---------|---------|--------|
10-
| GPU | 英伟达 | 全部 | Yes | Yes | Yes |
11-
| MLU | 寒武纪 | 370, 590 | Yes | Yes | No |
12-
| DCU | 中科海光 | Z100, Z100L | Yes | Yes | No |
8+
| 类型 | 制造商 | 型号 | 内存隔离 | 核心隔离 | 多卡支持 |
9+
|-----|-------|-------------------|---------|---------|--------|
10+
| GPU | 英伟达 | 全部 | Yes | Yes | Yes |
11+
| MLU | 寒武纪 | 370, 590 | Yes | Yes | No |
12+
| DCU | 中科海光 | 全部 | Yes | Yes | No |
1313
| NPU | 华为昇腾 | 910B, 910B3, 310P | Yes | Yes | No |
14-
| GPU | 天数智芯 | 全部 | Yes | Yes | No |
15-
| GPU | 摩尔线程 | MTT S4000 | Yes | Yes | No |
16-
| DPU | Teco | 检查中 | 进行中 | 进行中 | No |
14+
| GPU | 天数智芯 | 全部 | Yes | Yes | No |
15+
| GPU | 摩尔线程 | MTT S4000 | Yes | Yes | No |
16+
| DPU | Teco | 检查中 | 进行中 | 进行中 | No |

i18n/zh/docusaurus-plugin-content-docs/version-v2.5.1/userguide/device-supported.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ HAMi支持的设备视图如下表所示:
99
|-------------|------------|-------------|-----------|---------------|-------------------|
1010
| GPU | NVIDIA | 全部 | Yes | Yes | Yes |
1111
| MLU | Cambricon | 370, 590 | Yes | Yes | No |
12-
| DCU | Hygon | Z100, Z100L | Yes | Yes | No |
12+
| DCU | Hygon | 全部 | Yes | Yes | No |
1313
| Huawei Ascend | Huawei | 910B, 910B3, 310P | Yes | Yes | No |
1414
| GPU | iluvatar | 全部 | Yes | Yes | No |
1515
| GPU | Mthreads | MTT S4000 | Yes | Yes | No |

i18n/zh/docusaurus-plugin-content-docs/version-v2.6.0/faq/faq.md

Lines changed: 8 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -5,15 +5,15 @@ title: 常见问题
55

66
## 支持的设备厂商及具体型号
77

8-
| **GPU 厂商** | **GPU 型号** | **粒度** | **多 GPU 支持** |
9-
| --- | --- | --- | --- |
10-
| NVIDIA | 几乎所有主流消费级和数据中心 GPU | 核心 1%,显存 1M | 支持。多 GPU 仍可通过虚拟化进行拆分和共享。 |
8+
| **GPU 厂商** | **GPU 型号** | **粒度** | **多 GPU 支持** |
9+
| --- |-----------------------| --- | --- |
10+
| NVIDIA | 几乎所有主流消费级和数据中心 GPU | 核心 1%,显存 1M | 支持。多 GPU 仍可通过虚拟化进行拆分和共享。 |
1111
| 昇腾 | 910A、910B2、910B3、310P | 最小粒度取决于卡类型模板。参考[官方模板](https://www.hiascend.com/document/detail/zh/mindx-dl/50rc1/AVI/cpaug/cpaug_0005.html)| 支持,但当 `npu > 1` 时不支持拆分,整卡独占。 |
12-
| 海光 | Z100、Z100L、K100-AI | 核心 1%,显存 1M | 支持,但当 `dcu > 1` 时不支持拆分,整卡独占。 |
13-
| 寒武纪 | 370、590 | 核心 1%,显存 256M | 支持,但当 `mlu > 1` 时不支持拆分,整卡独占。 |
14-
| 天数智芯 | 全部 | 核心 1%,显存 256M | 支持,但当 `gpu > 1` 时不支持拆分,整卡独占。 |
15-
| 摩尔线程 | MTT S4000 | 核心为 1 个核心组,显存 512M | 支持,但当 `gpu > 1` 时不支持拆分,整卡独占。 |
16-
| 魅特思 | MXC500 | 不支持拆分,只能整卡分配。 | 支持,但所有分配均为整卡。 |
12+
| 海光 | 全部 | 核心 1%,显存 1M | 支持,但当 `dcu > 1` 时不支持拆分,整卡独占。 |
13+
| 寒武纪 | 370、590 | 核心 1%,显存 256M | 支持,但当 `mlu > 1` 时不支持拆分,整卡独占。 |
14+
| 天数智芯 | 全部 | 核心 1%,显存 256M | 支持,但当 `gpu > 1` 时不支持拆分,整卡独占。 |
15+
| 摩尔线程 | MTT S4000 | 核心为 1 个核心组,显存 512M | 支持,但当 `gpu > 1` 时不支持拆分,整卡独占。 |
16+
| 魅特思 | MXC500 | 不支持拆分,只能整卡分配。 | 支持,但所有分配均为整卡。 |
1717

1818
## 什么是 vGPU?为什么看到 10 个 vGPU 却无法在同一张卡上分配两个 vGPU?
1919

i18n/zh/docusaurus-plugin-content-docs/version-v2.6.0/userguide/device-supported.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@ HAMi支持的设备视图如下表所示:
99
|-------------|------------|-------------|-----------|---------------|-------------------|
1010
| GPU | NVIDIA | 全部 | Yes | Yes | Yes |
1111
| MLU | Cambricon | 370, 590 | Yes | Yes | No |
12-
| DCU | Hygon | Z100, Z100L | Yes | Yes | No |
12+
| DCU | Hygon | 全部 | Yes | Yes | No |
1313
| Huawei Ascend | Huawei | 910B, 910B3, 310P | Yes | Yes | No |
1414
| GPU | iluvatar | 全部 | Yes | Yes | No |
1515
| GPU | Mthreads | MTT S4000 | Yes | Yes | No |

0 commit comments

Comments
 (0)