noinlining in clang/llvm

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

noinlining in clang/llvm

Eric Fiselier via cfe-dev
Chandler Carruth at a CppCon/NWCPP talk said that Clang will not
respect the noinline gcc attribute (or clang attribute).

Sometimes it is needed for semantic correctness that a method not be
inlined (e.g. for GC lifetime visibility etc.)

Is this true?

Video: https://youtu.be/9RFYVbHRoLg?t=3822
_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: noinlining in clang/llvm

Eric Fiselier via cfe-dev
My understanding is that the portion of the talk you've linked to, is in the context of profiling hints (e.g. if profiling detects just a single callsite for a specific function, always inline its code, that sort of thing).

The noinline attribute works just fine.


- ½

On 2 October 2017 at 23:44, Hayden Livingston via cfe-dev <[hidden email]> wrote:
Chandler Carruth at a CppCon/NWCPP talk said that Clang will not
respect the noinline gcc attribute (or clang attribute).

Sometimes it is needed for semantic correctness that a method not be
inlined (e.g. for GC lifetime visibility etc.)

Is this true?

Video: https://youtu.be/9RFYVbHRoLg?t=3822
_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev


_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev