gaspersic/triton
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[documentation][triton-c] improved wording on Triton-C being

Browse Source 
single-threaded
This commit is contained in:
Philippe Tillet
2019-09-10 13:36:34 -04:00
parent 7f21a63ae1
commit 8111d56ee9
1 changed files with 1 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
docs/triton-c.md
View File

@@ -66,7 +66,6 @@ float v[16, 32] = u[:, newaxis]; // broadcasting along the second axis
```
which is valid _Triton-C_.
_Note: Triton-C is single-threaded._
- **Portability**: One other issue with our initial C99 program is that it is not portable. While it will run well on a single CPU thread, the operation `z = x + y` would perform poorly on a GPU Streaming Processor as it would execute on a single thread only. For this readon, it would have to be rewritten for GPUs as follows:
```
@@ -76,7 +75,7 @@ float x = 3.14;
float y = 5.17;
float z = x + y
```
In Triton-C, the same code can be used on many different platforms (only CPUs and GPUs are supported at the moment).
In Triton-C, the same code can be used on many different platforms (only CPUs and GPUs are supported at the moment). Furthermore, Triton-C is single-threaded, hence easier to write than CUDA.
- **Performance**: Another issue with our initial C99 code snippet is its performance. Although the loops are unrolled, the program does not carry the data-flow information of array operations. This issue gets more and more problematic as programs get increasingly complex, eventually culminating in matrix multiplication being remarkably hard to optimize.