## Sparc cache flush issues (obscure)

Larceny detects whether cache flushing is necessary on your machine. You can also control cache flushing behavior explicitly manually with a switch in Rts/Makefile. Do so only with extreme care. 

It has been observed that forcing cache flushes on some systems that do not have split caches, can cause run times to increase noticeably in programs that allocate a great deal (because cache flushing is performed during garbage collection). Run times increase both in the user and system code. For example, running on one particular machine with a non-hw or specialized flush instruction, the basic LL1 benchmark runs in 25.3+4.0 with cache flushing on and in 22.0+0.0 with cache flushing turned off (numbers for illustration only). 

The systems where this has so far been observed are: 

  * Sun 4/600 (Sparc-2 multiprocessor, 4 CPUs) SunOS 4.1.3_U1 

I am surprised that the cache flush is slow on this one; it may be that the IFLUSH means something special because it is a multiprocessor. 

* * *

_$Id$_  
[larceny@ccs.neu.edu][1]

   [1]: mailto:larceny@ccs.neu.edu