Re: [apache/tvm-rfcs] [RFC] Adding initial SVE implementation (#18)

[sjoerdmeijer]

@masahi, about: 
 
> If possible, I think it is better not to introduce user facing changes, since 
> as far as an API is concerned, `s[C].vectorize(...)` is already vector-length 
> agnostic.

which I think is very closely related to an earlier inline comment:

> As I commented above, I'd like to continue using s[C].vectorize(...) and when 
> the feature is available, enable SVE by a target attribute. So I don't expect 
> any user facing work.

I think we do need a user-facing option to toggle fixed/scalable vectorisation. 
If the vectorisation strategy is selected based on an available target 
attribute, we loose control to choose fixed/scalable vectorisation. For 
architectures that do support scalable vectorisation, fixed width might still 
be preferable in some cases.

I think this is similar to Clang's loop pragmas. For example, the 
`vectorise_width` pragma has been extended with an optional second argument 
`fixed|scalable`:

`vectorize_width(_value_[, fixed|scalable]),`

see also the Clang docs 
[here](https://clang.llvm.org/docs/LanguageExtensions.html#extensions-for-loop-hint-optimizations).

So I see two approaches:
- we extend `s[C].vectorize(...)` to take an optional fixed/scalable boolean 
value, similar to Clang's loop pragma, which defaults to fixed if omitted, 
- or introduce `s[C].vectorize_scalable(...)` as proposed in this RFC.

I personally don't have any preference. But now I am wondering if extending 
`s[C].vectorize(...)`, the first option, would be better. What do you think?

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/18#issuecomment-910235468

Re: [apache/tvm-rfcs] Additional Target Hooks RFC (#10)

[Mark Shields]

This LGTM but I'm thinking going back to relay_to_tir being a 
Function->PrimFunc packed func as you implemented in the companion PR is better 
since it makes it easier for all the caching, call rewriting, dynamic shape 
handling and other bookkeeping to be shared. Have you tried out the pass 
approach and convinced youself that reuse is still possible?

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/10#issuecomment-910485693

[apache/tvm-rfcs] RFC for TVM Documentation Refactoring (#27)

[Chris Hoge]

You can view, comment on, or merge this pull request online at:

  https://github.com/apache/tvm-rfcs/pull/27

-- Commit Summary --

  * RFC for TVM Documentation Refactoring

-- File Changes --

A rfcs/00xx-refactor-docs.md (430)

-- Patch Links --

https://github.com/apache/tvm-rfcs/pull/27.patch
https://github.com/apache/tvm-rfcs/pull/27.diff

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/27

Re: [apache/tvm-rfcs] [RFC] Adding initial SVE implementation (#18)

[masahi]

Thanks @sjoerdmeijer @giuseros, I didn't imagine that there would be a case 
where mixing fixed and scalable vectorization is beneficial. I prefer 
`s[C].vectorize(..., scalable=True)` to `s[C].vectorize_scalable(...)` but both 
seem fine.

Any other comments @tqchen @tkonolige?

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/18#issuecomment-910821675

Re: [apache/tvm-rfcs] RFC for TVM Documentation Refactoring (#27)

[黄若辰]

unsubsribe




On 09/01/2021 15:06, Chris Hoge wrote:

You can view, comment on, or merge this pull request online at:

 https://github.com/apache/tvm-rfcs/pull/27

-- Commit Summary --

 * RFC for TVM Documentation Refactoring

-- File Changes --

   A rfcs/00xx-refactor-docs.md (430)

-- Patch Links --

https://github.com/apache/tvm-rfcs/pull/27.patch
https://github.com/apache/tvm-rfcs/pull/27.diff

--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/27

Re: [apache/tvm-rfcs] RFC for TVM Documentation Refactoring (#27)

[黄若辰]

cancel




On 09/01/2021 16:56, 黄若辰 wrote:
unsubsribe




On 09/01/2021 15:06, Chris Hoge wrote:

You can view, comment on, or merge this pull request online at:

https://github.com/apache/tvm-rfcs/pull/27

-- Commit Summary --

* RFC for TVM Documentation Refactoring

-- File Changes --

  A rfcs/00xx-refactor-docs.md (430)

-- Patch Links --

https://github.com/apache/tvm-rfcs/pull/27.patch
https://github.com/apache/tvm-rfcs/pull/27.diff

--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/27

Re: [apache/tvm-rfcs] [RFC]PyTorchTVM (#25)

[Junru Shao]

@Meteorix Thanks for your responds! I just updated the comments according to 
your feedback. Don't hesitate to ping me if you need any inputs. Thanks a lot!

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/25#issuecomment-911064462

RE: [apache/tvm-rfcs] [RFC]PyTorchTVM (#25)

[ZHENG Jianwei]

Unsubscribe



Sent from Mail for Windows



From: Junru Shao
Sent: Thursday, September 2, 2021 10:40 AM
To: apache/tvm-rfcs
Cc: Subscribed
Subject: Re: [apache/tvm-rfcs] [RFC]PyTorchTVM (#25)



@Meteorix Thanks for your responds! I just updated the comments according to 
your feedback. Don't hesitate to ping me if you need any inputs. Thanks a lot!

--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://apc01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fapache%2Ftvm-rfcs%2Fpull%2F25%23issuecomment-911064462&data=04%7C01%7Cjianweizheng%40ust.hk%7C9d67ecd4c5c744ebb4d408d96dbaf6a9%7Cc917f3e2932249269bb3daca730413ca%7C1%7C0%7C637661472038884770%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=406fjhslENYl7mptwAHb4GEi0G08zCOf4BhMACA%2FopA%3D&reserved=0