Comments (5)
Yes, this is correct. For Triton we've been focusing mostly on server GPUs, so V100 and A100. It's probably not much work to get it to work on M60 and K80 though. I'll leave this open in case anyone wants to take up this issue :)
from triton.
Hi! It is possible for sure. Where are you getting the error specifically? You'd just need to edit the autotuner's config so it doesn't use anything that your hardware doesn't support
from triton.
@ptillet Thanks a lot. This issue happens at below code:
line 551 of /python/triton/code_gen.py
`tt_device = _triton.driver.cu_device(device.index, False)
# Compile to machine code
mod, ker, shared_mem, ir_asm = _triton.code_gen.add_passes_to_emit_bin(generator.module, tt_device, num_warps, num_stages)
if shared_mem > tt_device.max_shared_memory():
raise OutOfResources(shared_mem, tt_device.max_shared_memory(), "shared memory")
return Binary(mod, ker, num_warps, num_stages, shared_mem, ir_asm)`
I tried on M60&K80. tt_device.max_shared_memory() is 49152 and has this issue. For V100. seems like it's enough
from triton.
Hi, @ptillet
I find the following information from the readme. Seem that triton only supports GPU with Compute Capability >= 7.0?
The compute capability of M60 is 5.2 and K80 is 3.7
Compatibility
Supported Platforms:
- Linux
Supported Hardware:
- NVIDIA GPUs (Compute Capability 7.0+)
- Under development: AMD GPUs, CPUs
from triton.
I'm guessing K80 support is for Google Colab. They have been providing T4 more (anecdotally), last 4 notebooks I got were all T4 so maybe just asking for a new kernel if you get a K80 is workable. T4 worked just fine with Triton and got some modest speedups on the vector add tutorial at large sizes
from triton.
Related Issues (20)
- tl.cumsum seems emitting an internal error. HOT 1
- How to perform a store operation on a part of a Tensor? HOT 1
- Question regarding stride HOT 1
- [AMD] Undefined behavior sanitizer invalid-bool-load in optimize_epilogue.mlir HOT 5
- int8 x bfloat16 matmul tests fail on 4090s due to numerical error
- Incorrect result with threadsPerWarp of [2, 2, 8] for a thread block of [2, 2, 32] HOT 16
- M2 Mac Build from Source Failure: MLIR Configuration Error HOT 1
- RuntimeError: Triton Error [HIP]: Code: 1, Messsage: invalid argument
- StackTrace handler on python module does not allow signal to propagate. HOT 5
- Associative scan with non-scalar inputs
- how to use transpose in pytorch HOT 1
- Support masking in atomic_cas?
- Triton Error [CUDA]: device kernel image is invalid HOT 1
- Cannot specify which device to use HOT 2
- flash-attention gradient calculation fail due to numerical error HOT 1
- Hitting an assertion in `RemoveLayoutConversions` Pass. Relevant for both cuda and hip backends. HOT 2
- Help build WHL for Windows
- How to add a pow function in python.triton.language.core? HOT 1
- tensor not support index in nvidia gpu?
- No matching distribution found for triton-nightly
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from triton.