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: [PATCH] Fix CSE bogus RTL creation


On 08/05/2010 12:27 AM, Mark Mitchell wrote:
Steven Bosscher wrote:

I'm not opposing the patch at all costs, but definitely the bug hasn't
been analyzed thoroughly enough.

And apparently we're now completely stalled. Paul, are you still working on this? Is there anything you can file in bugzilla, at least, so this discussion will not get lost?

We do seem badly stuck.


But, I don't understand, at this point, what we can do to get unstuck.
I've just reread the thread and I don't see a clear direction about what
form of patch would be acceptable here.  Or, if we don't understand that
yet, what information we need next.

We need to know why the non-canonical note gets into an insn. This is a much more serious bug than a non-canonical note.


After that, or if that fails, I posted another patch in the thread and I would like to understand if this patch also fixes the bug.

Paolo


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