Hi, On Fri, 11 Jun 2004, Martin Pitt wrote:
> I contacted upstream about this and he had a totally different > opinion: gcc developers disagree, we had same problem with the kernel, e.g. see: http://marc.theaimsgroup.com/?l=linux-kernel&m=107475162200773&w=2 There is nothing broken with the compiler and changing the constraints of the memory operand to a "=m"/"m" pair is correctly understood by all versions of gcc. bye, Roman
