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] MIPS32 DSP intrinsics


Richard Henderson wrote:
On Thu, Jun 02, 2005 at 03:37:21PM -0700, Mark Mitchell wrote:

If they're just unspecs/intrinsics, then the instructions are magic, and as long as they're correctly conditionalized, what's to worry about?


Should they be using regular rtl patterns instead of unspecs?
Should they be using unspec-volatile instead of plain unspec?

Fair questions. But, what are the consequences of getting it wrong? I think, just that the compiler doesn't work as well as it might on the new chip.


Is this situation really unprecedented? Have we never had a port in the FSF tree without at least one of publicly available simulators, manuals, or silicon? I'm not asking facetiously; I'm honestly not sure. My impression is that we've had such ports in the past.

--
Mark Mitchell
CodeSourcery, LLC
mark@codesourcery.com
(916) 791-8304


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