github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-09 15:08:02 -05:00
Code Issues Packages Projects Releases Wiki Activity

docs: fix broken links and update is_floating_point (#6023)

Browse Source 
* docs: fix broken links and update is_floating_point

broken links would only show as INFO and not an error.

* make doc andhors warn
This commit is contained in:
chenyu
2024-08-10 15:58:48 -04:00
committed by GitHub
parent d58ae17771
commit d82370f6ef
5 changed files with 12 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docs/developer/hcq.md
View File

@@ -2,7 +2,7 @@
## Overview
The main aspect of HCQ-compatible runtimes is how they interact with devices. In HCQ, all interactions with devices occur in a hardware-friendly manner using [command queues](#commandqueues). This approach allows commands to be issued directly to devices, bypassing runtime overhead such as HIP or CUDA. Additionally, by using the HCQ API, these runtimes can benefit from various optimizations and features, including [HCQGraph](#hcqgraph) and built-in profiling capabilities.
The main aspect of HCQ-compatible runtimes is how they interact with devices. In HCQ, all interactions with devices occur in a hardware-friendly manner using [command queues](#command-queues). This approach allows commands to be issued directly to devices, bypassing runtime overhead such as HIP or CUDA. Additionally, by using the HCQ API, these runtimes can benefit from various optimizations and features, including [HCQGraph](#hcqgraph) and built-in profiling capabilities.
### Command Queues
@@ -97,7 +97,7 @@ Each HCQ-compatible device must allocate two signals for global synchronization
### HCQ Compatible Allocator
The `HCQAllocator` base class simplifies allocator logic by leveraging [command queues](#commandqueues) abstractions. This class efficiently handles copy and transfer operations, leaving only the alloc and free functions to be implemented by individual backends.
The `HCQAllocator` base class simplifies allocator logic by leveraging [command queues](#command-queues) abstractions. This class efficiently handles copy and transfer operations, leaving only the alloc and free functions to be implemented by individual backends.
::: tinygrad.device.HCQAllocator
options:

2
docs/developer/runtime.md
View File

@@ -4,7 +4,7 @@
A typical runtime consists of the following parts:
- [Compiled](#device)
- [Compiled](#compiled)
- [Allocator](#allocator)
- [Program](#program)
- [Compiler](#compiler)

2
docs/index.md
View File

@@ -40,7 +40,7 @@ In tinygrad, you can do [`x.conv2d(w, b)`](tensor/ops.md/#tinygrad.Tensor.conv2d
### tinygrad is lazy
When you do `a+b` in tinygrad, nothing happens. It's not until you [`realize`](tensor/index.md/#tinygrad.Tensor.realize) the Tensor that the computation actually runs.
When you do `a+b` in tinygrad, nothing happens. It's not until you [`realize`](tensor/properties.md#tinygrad.Tensor.realize) the Tensor that the computation actually runs.
### tinygrad requires @TinyJit to be fast