Re: memory_object_lock_request and memory_object_data_return fnord

[Roland McGrath]

> I was thinking along the lines of changing the __memcpy's found in
> libc/sysdep/i386/{,i[456]86}/memcpy.S et al. to __real_memcpy and then
> having a libc/sysdep/mach/memcpy.c do:

Eh, that's only half of it.  memcpy is as likely to be inlined in the
bits/string.h magic, or by the compiler.  And then there's doing this all
over again for every other platform, and keeping track of new implementations.
And even if you hack all the header files, there's still inlined versions
from things compiled for GNU/Linux one day when we have binary compatibility.

I am very dubious about changing memcpy.