This is the mail archive of the 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]

[PATCH] Document experimental nature of -fnew-ra?


As discussed on the bugmaster list, currently the
new-regalloc branch has diverged considerably from
3.4-branch/mainline wrt the new ra. These changes seem
unlikely to be merged back before 3.4.0, and there is
so far no discussion of when they will make it onto

Given that we currently have 15 or so bugs against
-fnew-ra on 3.4-branch, many of which may be fixed on
the new-regalloc branch, and there currently is little
progress on fixing them for 3.4/mainline, I think it
makes sense to document that -fnew-ra, like -fssa (of
rtl ssa days) is still very much experimental.

Also, because bugs reported against -fnew-ra on
3.4-branch are not necessarily still present against
new-regalloc branch, I've suspended them until such a
time as they are verified on the new-regalloc branch,
or new-regalloc is merged back to mainline.

Does the attached patch look reasonable for
mainline/3.4? Thanks,


Do you Yahoo!?
Yahoo! SiteBuilder - Free web site building tool. Try it!

Attachment: fnew-ra-doc.patch
Description: fnew-ra-doc.patch

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