Thanks @areusch for your response!
> if TVMScript is a core way in which TIR is used, I'd argue we should treat
> them conceptually as joined (e.g. TVMScript as the recommended roundtrip text
> format for TIR). What are your thoughts there?
Let's phrase it this way: TVMScript serves as a fronte
Merged #70 into main.
--
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm-rfcs/pull/70#event-6781982914
You are receiving this because you are subscribed to this thread.
Message ID:
> people use text format for readability, and in most usecases, go for json
> reliable serialization.
what do you mean by "reliable" here? if they're truly roundtrippable, aren't
they both reliable? just trying to understand :)
> Instead, for metaprogramming capability, one may need to interlea
> How would we enforce this fixed global symbol? Also from name supply?
When a function need to fix a global name, it will attach an attr
"global_symbol" with the name. That attr can be used say by BYOC to pick a
name, attach the global symbol, so followup passes respect that constraint.
We c
[quote="mbs-octoml, post:6, topic:12944"]
Other than the plumbing, is there an issue with threading a name supply so that
globals have a unique and appropriately hinted name at birth?
[/quote]
One question in my mind is: there are some names which must be stable e.g.
`__main__`, `get_c_metadat
we discussed this at the June 6 [community
meeting](https://discuss.tvm.apache.org/t/next-tvm-community-meeting-june-8-2022/12900).
a significant chunk of the meeting was spent presenting the RFC, and we had
about 15 minutes of discussion at the end.
i think there is more to be discussed here.
The recording has been [posted](https://youtu.be/1fSfhgVp0ms).
---
[Visit
Topic](https://discuss.tvm.apache.org/t/next-tvm-community-meeting-june-8-2022/12900/3)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here](http
[quote="mbs-octoml, post:6, topic:12944"]
(While doing this we could also ensure names begin in GlobalVar form instead of
String form, since the latter causes multiple GlobalVars to be created with the
same name
[/quote]
Indeed the `name_hint` field in GlobalVar is supposed to work for that
p
Other than the plumbing, is there an issue with threading a name supply so that
globals have a unique and appropriately hinted name at birth? It's not too hard
to support name supply splitting such that names can be drawn from independent
supplies without collision. It is also possible to refi
i agree with that direction, but I think to do that requires us to merge the
IRModules so that mangling transforms are applied globally rather than to just
a part of the program. do you agree?
---
[Visit
Topic](https://discuss.tvm.apache.org/t/pre-rfc-name-mangling-in-irmodules/12944/5)
> > > @leandron, looking at Docker Hub (https://hub.docker.com/_/hello-world)
> > > it would appear the convention for image names is to use `-` there as
> > > well (i.e. `tlcpack/ci-cpu` rather than `tlcpack/ci_cpu`) can we go for
> > > that one?
> >
> >
> > Sure. I’ll push an updated version
Agree that mangling should be deferred as long as possible. I think the keys is
not to push all mangling to a certain stage, but have a clear spec about what
can be mangled, and what cannot
---
[Visit
Topic](https://discuss.tvm.apache.org/t/pre-rfc-name-mangling-in-irmodules/12944/4)
to
Please join us to welcome [@apeskov](https://github.com/apeskov) as a new
reviewer to TVM.
Alexander has contributed to the BNNS BYOC backend, and iOS support, and has
ongoing work on improving DNNL support via BYOC.
* [Commits History](https://github.com/apache/tvm/commits?author=apeskov)
*
Please join us to welcome [@vvchernov](https://github.com/vvchernov) as a new
reviewer to TVM.
Valery has made contributions to improve support for LSTMs and speech synthesis
models like Tacotron2 in TVM. In addition he’s made improvements to the Relay
Virtual Machine to reduce input/output c
One related thing was that we intended to consolidate all of the different
functions into one IRModule (e.g. those passed to external codegen with those
passed to ordinary TVM codegen). If we had done that, we could probably do the
following:
- If a Relay-to-Runtime hook is used, find a way t
This is somewhat related to linkage itself.
Specifically, depending on the progress of lowering, there can be a need to
decide on name earlier if some of the external codegen involved(e.g. BYOC that
pre-decides on "global_symbol" of the name, or user specifies that they would
like to "lock"
Hi all,
In [[mini-RFC] Name mangling in
AOT](https://discuss.tvm.apache.org/t/mini-rfc-name-mangling-in-aot/9907), we
discussed ways to accommodate the `runtime::Module` tree in an embedded
environment, where we prefer to call functions directly in the generated C
using their symbol name rat
In the spirit of keeping this simple to review, the text of this RFC has been
reformulated to cover the `target_parser` replacement for `preprocessor` which
@junrushao1994 proposed - I've raised a follow on for the `features` additional
to the `Target` so that can also be reviewed in isolation w
You can view, comment on, or merge this pull request online at:
https://github.com/apache/tvm-rfcs/pull/78
-- Commit Summary --
* Add Target Features RFC
-- File Changes --
A rfcs/0078-target-features.md (178)
-- Patch Links --
https://github.com/apache/tvm-rfcs/pull/78.patch
https:
I think it is an exciting RFC! So has it merged into TVM main stream? And can
you share a symbolic shape example with TE? Thank you
---
[Visit
Topic](https://discuss.tvm.apache.org/t/rfc-dietcode-an-auto-scheduler-for-dynamic-tensor-programs/12757/5)
to respond.
You are receiving this be
20 matches
Mail list logo
