Fwd: Fw: N4456 (small string optimisation)

61 views
Skip to first unread message

Michael Wong

unread,
Apr 16, 2015, 10:19:54 AM4/16/15
to unofficial-r...@googlegroups.com
I think I have gotten almost as much query on this paper's publication as TM. Here is someone who wants to help with SSO.

---------- Forwarded message ----------
From: Michael Wong <mich...@ca.ibm.com>
Date: Thu, Apr 16, 2015 at 10:18 AM
Subject: Fw: N4456 (small string optimisation)
To: Michael Wong <fragga...@gmail.com>




_________________________________________________________
Regards, Michael
ISOCPP.org Director, VP http://isocpp.org/wiki/faq/wg21#michael-wong
OpenMP CEO: http://openmp.org/wp/about-openmp/
My Blogs: http://ibm.co/pCvPHR
C++11 status: http://tinyurl.com/43y8xgf
Boost test results http://www.ibm.com/support/docview.wss?rs=2239&context=SSJT9L&uid=swg27006911
C/C++ Compilers Feature Request Page http://www.ibm.com/developerworks/rfe/?PROD_ID=700
Chair of WG21 SG5 Transactional Memory: https://groups.google.com/a/isocpp.org/forum/?hl=en&fromgroups#!forum/tm

IBM Corporation
XL C++ Compiler kernel Development
IBM z Systems Software,IBM Systems Unit
IBM Canada Ltd., C2/KD2/8200/MKM
8200 Warden Avenue
Markham, Ontario L6G 1C7
W:905-413-3283 F:905-413-4839
OpenMPCon 2015


----- Forwarded by Michael Wong/Toronto/IBM on 04/16/2015 10:18 AM -----

From: Elliot Goodrich <elliotg...@gmail.com>
To: Michael Wong/Toronto/IBM@IBMCA
Date: 04/16/2015 09:21 AM
Subject: Re: N4456 (small string optimisation)




Thank you for replying. I'm glad it can help and feel free to use any of it in the paper.

Kind regards,
Elliot

On 16 April 2015 at 13:48, Michael Wong <mich...@ca.ibm.com> wrote:


Sean Middleditch

unread,
Apr 16, 2015, 4:51:55 PM4/16/15
to Michael Wong, unofficial-r...@googlegroups.com
Looks like a cool implementation.

By itself, it's the kind of thing we'd provision and put into a non-standard engine-specific string. I don't think the standard needs to mandate any implementation, but if it guaranteed some reliable and reasonable minimum and offered a good compile-time way to assert on the implementation-defined minimum, we might be in good shape. The SSO-23 work if nothing else gives strong evidence as to what a reasonable standardized minimum should be.

I've seen more complaints about unreliable SSO in std::function than std::string, to be honest. Especially as std::function is a key modern C++ facility used in various patterns built around avoiding exceptions, not to mention event-driven gameplay systems.

On the string front, first-class support for string_view is more useful than anything else, because there tends not to be a lot of mutable strings being passed around in games or many embedded apps. That said, when you _do_ need std::string, reliable SSO sizes is handy.

I think if we summarized N4456, it would come down to "give the developer as much control as they want."

--
You received this message because you are subscribed to the Google Groups "unofficial-real-time-cxx" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unofficial-real-ti...@googlegroups.com.
To post to this group, send email to unofficial-r...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/unofficial-real-time-cxx/CAAUWkG8ja4pjzYBy06qa1hcgOv%3De6AKo7vi-7wC%3D%3DicW_jr2tQ%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.



--
Sean Middleditch
http://seanmiddleditch.com
Reply all
Reply to author
Forward
0 new messages