You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2021/01/13 20:14:40 UTC
[GitHub] [tvm] anijain2305 opened a new pull request #7273: [CUDA] [Codegen] Ensuring atleast one thread block for dynamism
anijain2305 opened a new pull request #7273:
URL: https://github.com/apache/tvm/pull/7273
Topk was failing on CUDA when k is a var and its value is 0 at runtime. At closer inspection I found that there are 0 thread blocks at runtime. This PR ensures that there is atleast 1 thread block.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] anijain2305 edited a comment on pull request #7273: [CUDA] [Codegen] Ensuring atleast one thread block to handle empty tensor
Posted by GitBox <gi...@apache.org>.
anijain2305 edited a comment on pull request #7273:
URL: https://github.com/apache/tvm/pull/7273#issuecomment-759765797
> hmm, I think I've already added a fix for such cases, here:
>
> https://github.com/apache/tvm/blob/82942fb33fd6e3572897c815af16905c4f75c2a4/python/tvm/tir/ir_builder.py#L205-L206
>
> Do you know why it is not working? cc @mbrookhart
Is this because this the lines that you suggested are specific to IR Builder, while the failure that I see is for injective schedule? My failures was coming for an injective schedule.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] masahi merged pull request #7273: [CUDA] [Codegen] Ensuring atleast one thread block to handle empty tensor
Posted by GitBox <gi...@apache.org>.
masahi merged pull request #7273:
URL: https://github.com/apache/tvm/pull/7273
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] masahi commented on pull request #7273: [CUDA] [Codegen] Ensuring atleast one thread block to handle empty tensor
Posted by GitBox <gi...@apache.org>.
masahi commented on pull request #7273:
URL: https://github.com/apache/tvm/pull/7273#issuecomment-759763585
hmm, I think I've already added a fix for such cases, here:
https://github.com/apache/tvm/blob/82942fb33fd6e3572897c815af16905c4f75c2a4/python/tvm/tir/ir_builder.py#L205-L206
Do you know why it is not working? cc @mbrookhart
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] mbrookhart commented on pull request #7273: [CUDA] [Codegen] Ensuring atleast one thread block to handle empty tensor
Posted by GitBox <gi...@apache.org>.
mbrookhart commented on pull request #7273:
URL: https://github.com/apache/tvm/pull/7273#issuecomment-759768625
Yeah, I think this change catches it at a lower level. We might not need the ir_builder change after this.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] anijain2305 commented on pull request #7273: [CUDA] [Codegen] Ensuring atleast one thread block for dynamism
Posted by GitBox <gi...@apache.org>.
anijain2305 commented on pull request #7273:
URL: https://github.com/apache/tvm/pull/7273#issuecomment-759723834
@kevinthesun @masahi @mbrookhart @zhiics @trevor-m Please review.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] anijain2305 commented on pull request #7273: [CUDA] [Codegen] Ensuring atleast one thread block to handle empty tensor
Posted by GitBox <gi...@apache.org>.
anijain2305 commented on pull request #7273:
URL: https://github.com/apache/tvm/pull/7273#issuecomment-759765797
> hmm, I think I've already added a fix for such cases, here:
>
> https://github.com/apache/tvm/blob/82942fb33fd6e3572897c815af16905c4f75c2a4/python/tvm/tir/ir_builder.py#L205-L206
>
> Do you know why it is not working? cc @mbrookhart
Is this because this the lines that you suggested are specific to IR Builder, while the failure that I see is for injective schedule?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org