This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
RE: Is fastcall broken?
- From: "John (Eljay) Love-Jensen" <eljay at adobe dot com>
- To: "Andrew Haley" <aph-gcc at littlepinkcloud dot COM>, "Daniel Lohmann" <daniel dot lohmann at informatik dot uni-erlangen dot de>
- Cc: "Angus" <angus at uducat dot com>, <gcc-help at gcc dot gnu dot org>
- Date: Sat, 30 Jun 2007 05:20:32 -0700
- Subject: RE: Is fastcall broken?
- References: <18054.14291.118304.751789@zebedee.pink>
Hi Andrew & Daniel & everyone,
Daniel says...
> Attributes that change a function to a non-standard calling convention effectively modify the interface of the function, which should be encoded into the (mangled) symbol name. Thereby incompatible prototypes on on the caller and callee side could be detected at link-time.
That is an excellent idea for C++ name-mangled identifiers!
Andrew says...
>Otherwise it's not such a bad idea, but it is an ABI change: such changes, being non-backwards compatible, are usually unpopular.
Arrgh. ABI change. Code kryptonite. :-(
Next time (if there is a next time) the ABI goes through a non-backwards compatible modification, it would be very good to incorporate Daniel's suggestion.
I presume if the "nothrow" attribute is approved by the C++ standards committee, the ABI would have to be revved. So that would be one such possible perhaps opportune time.
Sincerely,
--Eljay