[cfe-dev] libc++ and large stack frames

[Duncan P. N. Exon Smith via cfe-dev]

> On May 31, 2018, at 15:35, Richard Smith via cfe-dev <cfe-dev at lists.llvm.org> wrote:
> 
> On 31 May 2018 at 14:38, Friedman, Eli via cfe-dev <cfe-dev at lists.llvm.org <mailto:cfe-dev at lists.llvm.org>> wrote:
> On 5/31/2018 1:07 PM, Erik van der Poel via cfe-dev wrote:
> This is a proposal to add a configuration option to disable _LIBCPP_INLINE_VISIBILITY, which causes large stack frames in non-optimized builds because it uses the always_inline attribute to force significant amounts of inlining of libc++ code.
> 
> The new configuration option could be called _LIBCPP_DISABLE_INLINE_VISIBILITY.
> 
> Note that _LIBCPP_INLINE_VISIBILITY and _LIBCPP_ALWAYS_INLINE have identical definitions. One could be renamed to the other or both could be renamed to _LIBCPP_HIDE_FROM_ABI as part of this work.

I think it's useful to have different names for these, since they have different intent.  It would be reasonable to change the definition of one without changing the other.

Renaming _LIBCPP_INLINE_VISIBILITY to _LIBCPP_HIDE_FROM_ABI makes sense to me though.

One option we'd like to investigate (but haven't yet) is changing _LIBCPP_INLINE_VISIBILITY/_LIBCPP_HIDE_FROM_ABI to drop the "always_inline" but keep the "hidden".  Have you considered _LIBCPP_DISABLE_ALWAYS_INLINE_IN_INLINE_VISIBILITY (or equivalent)?  If not, why not?

> Alternatives considered and rejected:
> 
> 1. Remove __attribute__((__always_inline__)) from all include/__config macros.
> 2. Add _LIBCPP_DISABLE_ALWAYS_INLINE to control whether or not __always_inline__ is included in any __config macro.
> 3. Use the existing _LIBCPP_DISABLE_VISIBILITY_ANNOTATIONS to control both __visibility__ and __always_inline__.
> 4. Use the existing _LIBCPP_ABI_UNSTABLE to control _LIBCPP_INLINE_VISIBILITY.
> 
> 5. Start using __attribute__((internal_linkage)) instead, so a configuration option isn't necessary.  See http://lists.llvm.org/pipermail/cfe-dev/2016-October/051151.html <http://lists.llvm.org/pipermail/cfe-dev/2016-October/051151.html> .
> 
> Yes, that'd solve the problem, but it would also result in duplicate definitions for all the INLINE_VISIBILITY member functions that don't get inlined. As a consequence, __attribute__((internal_linkage)) would be a worse solution than the one being proposed here for those who don't care about the ABI stability guarantees that libc++ is trying to provide. Replacing the current use of __always_inline__ with __internal_linkage__ seems like it would probably be an improvement to me (conditional on whether the effects on code size are acceptable), but I think the change Erik is proposing and that one are largely orthogonal.

I agree.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20180604/ab146fa2/attachment.html>