This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: An unusual Performance approach using Synthetic registers
- From: Tom Lord <lord at emf dot net>
- To: dewar at gnat dot com
- Cc: denisc at overta dot ru, gcc at gcc dot gnu dot org, ja_walker at earthlink dot net
- Date: Sun, 5 Jan 2003 05:59:39 -0800 (PST)
- Subject: Re: An unusual Performance approach using Synthetic registers
- References: <20030105134518.DD9FEF2D5D@nile.gnat.com>
If you decide to allocate a base register, as was proposed at
one point,
Bah. I missed that (misread while skimming, actually). Maybe his
implementation approach is bogus after all. Why not use the FP, if
it's there, or SP when FP is omitted.
Remember again that the code to access SR's can be no better
than the code we generate right now for all accesses to local
variables, function arguments in memory, and spilled registers.
But it can also improve both locality and the temporally proximate
re-use of memory locations.
ja_walker is right: it's a worthwhile emperical question. And I'd
still bet you a beer or something that in 8 years, the answer to the
question is "yes" for the commodity systems. Have you ever had an
8-year old beer? Have you?!?!?
-t