[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [MIT-Scheme-devel] [commit f372dae] Don't call OS_free_pages after G
Taylor R Campbell
Re: [MIT-Scheme-devel] [commit f372dae] Don't call OS_free_pages after GC flip. We don't resize the heap and we will be reusing it.
Fri, 7 Oct 2011 19:29:47 +0000
IMAIL/1.21; Edwin/3.116; MIT-Scheme/9.1
Date: Fri, 7 Oct 2011 12:08:09 -0700
From: Joe Marshall <address@hidden>
> On Wed, Oct 5, 2011 at 7:50 PM, Taylor R Campbell <address@hidden>
> > Contrariwise, if I have six Scheme processes that have not freed their
> > unused pages and the OS is constantly swapping, then the whole world
> > gets much, much slower, especially at larger heap sizes.
Were they all actively computing, or were some of the processes waiting for
Some were waiting for input.
It may be the case that if they were all computing, then not doing the
extra work of OS_free_pages would make the Scheme computations run a
little faster. They would also run a little faster if the OS weren't
running any other processes, and they would also run a little faster
if they didn't poll for interrupts. But then my OS would be unusable
and Scheme would be interactively unresponsive.
The purpose of calling OS_free_pages is to make Scheme behave better
as a citizen of OSland and avoid bogarting more physical memory than
it needs, not necessarily to speed up Scheme computations.