This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: ping: back end reinitialization hooks to support mixed mips16/nomips16 compilation in same file


Richard Guenther wrote:
On 7/9/07, Mark Mitchell <mark@codesourcery.com> wrote:
So, I think the target-independent part of the patch is OK, but please
wait at least 72 hours to allow for objections.  If there are some, then
we'll need to work through the issues.  The MIPS changes should be
reviewed by a MIPS maintainer.

Please specify where you bootstrapped and tested this patch on and make sure you also cover a non-mips primary target.

Actually, this patch has been only lightly tested so far, and I'm doubtful that it's ready to commit even if it passes bootstrap validation. I posted it in the hopes of getting some feedback and discussion from the community before putting a lot more work into it. In particular, if the overwhelming consensus turned out to be that re-initializing the back end is just a Dumb Idea, I didn't want to spend weeks polishing and testing the implementation. OTOH, if the consensus is that it's not a Dumb Idea, is the way this patch goes about doing it the right approach? The set of things that are reinitialized is fairly specific to the needs of the MIPS back end; should we be striving for greater generality at the expense of adding more complication? Are there more hooks or initialization functions that should be added? Being more of a front-end hacker up until now ;-) I'm feeling at somewhat of a disadvantage in trying to engineer this restructuring of the back end, so any thoughts about the Right Thing To Do would be appreciated.


-Sandra


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]