[native-client-discuss] Clarification of EOL announcement refs - Nacl exception for extensions?

[Matthew Powers]

Hi list-

I had questions concerning the EOL plans - the two key references I know of
are these:

https://blog.chromium.org/2017/05/goodbye-pnacl-hello-webassembly.html

- We will remove support for PNaCl in the first quarter of 2018
everywhere except inside Chrome Apps and Extensions.

https://developer.chrome.com/native-client/migration

- Given the momentum of cross-browser WebAssembly support, we
plan to focus our native code efforts on WebAssembly going forward and plan
to remove support for PNaCl in Q1 2018 *(except for Chrome Apps*). We
believe that the vibrant ecosystem around *WebAssembly*
<http://webassembly.org/> makes it a better fit for new and existing
high-performance web apps and that usage of PNaCl is sufficiently low to
warrant deprecation.

When comparing these statements, I see that Nacl support for extensions
will continue to work is specifically called out in one reference and the
other reference says only Chrome Apps will continue to support Nacl.


Does anyone know for sure - is the second blog simply overlooking Extension
usage and neglected to mention it?


thanks,


Matt

[Soeren Balko]

I would also be interested to hear an updated statement from Google on
their plans to remove PNaCl from Chrome.

AFAICS, WebAssembly has not yet caught up with PNaCl in terms of
performance (no support for SIMD intrinsics and multithreading yet, some
Pepper APIs such as video decoder/encoder are also not available in
WebAssembly, etc.). Also, Chrome's support for WebAssembly seems to suffer
from teething problems re stability under modest memory pressure ("Ah
snap"). That being the case, we plan to move to running our PNaCl module
inside a "companion" browser extension. At least that is until WebAssembly
truly delivers in terms of performance, features, and stability. It would
thus be great to get an up-to-date statement from Google, addressing the
following concerns:

(1) Is the plan still to retire PNaCl from Chrome (except for browser
extensions) in Q1/2018? If so, which Chrome release is going to be first
affected? Or are these plans on hold while Chrome's WebAssembly support is
still behind PNaCl?
(2) When will Chrome's WebAssembly implementation support SIMD intrinsics
and multithreading?
(3) Will PNaCl indeed continue to run inside Chrome extensions?

Post by Matthew Powers
Hi list-
I had questions concerning the EOL plans - the two key references I know
https://blog.chromium.org/2017/05/goodbye-pnacl-hello-webassembly.html

- We will remove support for PNaCl in the first quarter of 2018
everywhere except inside Chrome Apps and Extensions.

https://developer.chrome.com/native-client/migration

- Given the momentum of cross-browser WebAssembly support, we
plan to focus our native code efforts on WebAssembly going forward and plan
to remove support for PNaCl in Q1 2018 *(except for Chrome Apps*). We
believe that the vibrant ecosystem around *WebAssembly*
<http://webassembly.org/> makes it a better fit for new and existing
high-performance web apps and that usage of PNaCl is sufficiently low to
warrant deprecation.

When comparing these statements, I see that Nacl support for extensions
will continue to work is specifically called out in one reference and the
other reference says only Chrome Apps will continue to support Nacl.
Does anyone know for sure - is the second blog simply overlooking
Extension usage and neglected to mention it?
thanks,
Matt

[v***@gmail.com]

I am wondering the same thing. The performance of PNaCl is unmatched
currently and we ported our app to WASM but its not performant enough.

Post by Soeren Balko
I would also be interested to hear an updated statement from Google on
their plans to remove PNaCl from Chrome.
AFAICS, WebAssembly has not yet caught up with PNaCl in terms of
performance (no support for SIMD intrinsics and multithreading yet, some
Pepper APIs such as video decoder/encoder are also not available in
WebAssembly, etc.). Also, Chrome's support for WebAssembly seems to suffer
from teething problems re stability under modest memory pressure ("Ah
snap"). That being the case, we plan to move to running our PNaCl module
inside a "companion" browser extension. At least that is until WebAssembly
truly delivers in terms of performance, features, and stability. It would
thus be great to get an up-to-date statement from Google, addressing the
(1) Is the plan still to retire PNaCl from Chrome (except for browser
extensions) in Q1/2018? If so, which Chrome release is going to be first
affected? Or are these plans on hold while Chrome's WebAssembly support is
still behind PNaCl?
(2) When will Chrome's WebAssembly implementation support SIMD intrinsics
and multithreading?
(3) Will PNaCl indeed continue to run inside Chrome extensions?

Post by Matthew Powers
Hi list-
I had questions concerning the EOL plans - the two key references I know
https://blog.chromium.org/2017/05/goodbye-pnacl-hello-webassembly.html

- We will remove support for PNaCl in the first quarter of
2018 everywhere except inside Chrome Apps and Extensions.

https://developer.chrome.com/native-client/migration

- Given the momentum of cross-browser WebAssembly support, we
plan to focus our native code efforts on WebAssembly going forward and plan
to remove support for PNaCl in Q1 2018 *(except for Chrome Apps*). We
believe that the vibrant ecosystem around *WebAssembly*
<http://webassembly.org/> makes it a better fit for new and existing
high-performance web apps and that usage of PNaCl is sufficiently low to
warrant deprecation.

When comparing these statements, I see that Nacl support for extensions
will continue to work is specifically called out in one reference and the
other reference says only Chrome Apps will continue to support Nacl.
Does anyone know for sure - is the second blog simply overlooking
Extension usage and neglected to mention it?
thanks,
Matt