On Wed, Aug 9, 2017 at 8:05 AM, Arjan van de Ven <ar...@linux.intel.com> wrote:
> On 8/9/2017 8:04 AM, Andi Kleen wrote:
>> I would add a new option
>> -fforce-frame-pointer
>> that gives the old -fno-omit-frame-pointer back, so that
>> users relying on frame pointers everywhere have a workaround.

This must be much more specific.  How does it impact:

1. LTO
2. Function inlining.
3. Partial function inlining.
4. Shrink-wrapping.

Any of them can impact function stack frame.

> that function should also fix the current situation where the framepointer
> is not useful
> because all actual code was moved to be outside the frame pointer code
> (e.g.  push/mov/pop followed by the actual function code)


Reply via email to