This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Complex MEM adjust_address instead of subreg (PR middle-end/21742)
On Wed, Jun 29, 2005 at 05:53:13PM -0400, David Edelsohn wrote:
> How can the mode-dependent restriction be addressed?
By performing the transformation and testing to see whether
the resulting address is valid.
Of course, this does mean that in the context of complex move
expansion we must have a fallback in case the new address isn't
valid. Which will always be true on ia64.
Which means that the easiest solution is your first patch.
But we should update the comments to include all this fine