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: altivec fix for starting_frame_offset


>>>>> Richard Henderson writes:

>> If we go down the currently proposed path, I am more and more
>> inclined to make this approach conditional on TARGET_ALTIVEC_ABI as
>> Aldy originally proposed.

Richard> I'm against this.  This stuff is already confusing.  Adding a 
Richard> conditional when both sides are _supposed_ to work out the same
Richard> will only compound the confusion.

	I'm not just thinking about the 3.2 trunk, but the 3.1 branch.  I
am willing to accept a correct version of this patch for the trunk, but I
think radical surgery to parameter passing and ABI compliance on the
branch is too risky.  If the patch only affected Altivec, I would be less
concerned.

David


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