Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

#0: Typo fix in TT distributed tech report #16308

Merged
merged 1 commit into from
Dec 24, 2024
Merged

#0: Typo fix in TT distributed tech report #16308

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 9 additions & 9 deletions tech_reports/TT-Distributed/TT-Distributed-Architecture-1219.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1016,18 +1016,18 @@ Below, we include snippets from both the TT-Mesh and TT-Metal examples to illust
MeshConfig mesh_config_0 = MeshConfig{.shape = virtual_mesh_shape, .offset = {0, 0}, .type=mesh_type};
MeshConfig mesh_config_1 = MeshConfig{.shape = virtual_mesh_shape, .offset = {0, 4}, .type=mesh_type};

DeviceHandle virtual_mesh_0 = CreateMeshDevice(mesh_config_0, 2 /* num_cqs */, DEFAULT_L1_SMALL_SIZE, DEFAULT_TRACE_REGION_SIZE, 1 /* num_command_queues */);
DeviceHandle virtual_mesh_1 = CreateMeshDevice(mesh_config_1, 2 /* num_cqs */, DEFAULT_L1_SMALL_SIZE, DEFAULT_TRACE_REGION_SIZE, 1 /* num_command_queues */);
DeviceHandle virtual_mesh_0 = CreateMeshDevice(mesh_config_0, DEFAULT_L1_SMALL_SIZE, DEFAULT_TRACE_REGION_SIZE, 1 /* num_command_queues */);
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

num_cqs and num_command_queues were repeated. This version also aligns with the existing mesh device impl.

DeviceHandle virtual_mesh_1 = CreateMeshDevice(mesh_config_1, DEFAULT_L1_SMALL_SIZE, DEFAULT_TRACE_REGION_SIZE, 1 /* num_command_queues */);
```

*Directly create raw handles to two Devices.*

```cpp
DeviceHandle device_0 = CreateDevice(
0, /* device_id */
2, /* num_hw_cqs */
DEFAULT_L1_SMALL_SIZE,
DEFAULT_TRACE_REGION_SIZE);
DeviceHandle device_0 = CreateDevice(
0, /* device_id */
2, /* num_hw_cqs */
DEFAULT_L1_SMALL_SIZE,
DEFAULT_TRACE_REGION_SIZE);
DeviceHandle device_1 = CreateDevice(
1, /* device_id */
2, /* num_hw_cqs */
Expand Down Expand Up @@ -1070,7 +1070,7 @@ DeviceLocalLayoutConfig per_device_buffer_config {
ShardedBufferConfig distributed_buffer_config_virtual_mesh_0 {
.mesh_device = virtual_mesh_0;
.buffer_type = BufferType::DRAM,
.global_tensor_shape = global_tensor_shape,
.global_buffer_shape = global_buffer_shape,
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

As per the declaration earlier in the document.

.distributed_shard_shape = device_shard_shape,
.global_buffer_size = distributed_buffer_size,
.device_shard_layout = per_device_buffer_config
Expand All @@ -1079,7 +1079,7 @@ ShardedBufferConfig distributed_buffer_config_virtual_mesh_0 {
ShardedBufferConfig distributed_buffer_config_virtual_mesh_1 {
.mesh_device = virtual_mesh_1;
.buffer_type = BufferType::DRAM,
.global_tensor_shape = global_tensor_shape,
.global_buffer_shape = global_buffer_shape,
.distributed_shard_shape = device_shard_shape,
.global_buffer_size = distributed_buffer_size,
.device_shard_layout = per_device_buffer_config
Expand Down
Loading