Sujet : Re: stacks are not hard, The joy of FORTRAN-like languages
De : lynn (at) *nospam* garlic.com (Lynn Wheeler)
Groupes : alt.folklore.computers comp.os.linux.miscDate : 05. Oct 2024, 05:07:09
Autres entêtes
Organisation : Wheeler&Wheeler
Message-ID : <87ed4vgsde.fsf@localhost>
References : 1 2 3 4 5 6
User-Agent : Gnus/5.13 (Gnus v5.13)
Lawrence D'Oliveiro <
ldo@nz.invalid> writes:
Soon after dynamic memory allocation was invented, it was discovered that
keeping lookaside lists of free blocks in common sizes speeded up
allocations immensely.
>
This was all well-known by about the 1980s, if not before.
implemented in CP67 kernal ("subpools") code 1970 (released CP67
R3) ... following from presentation I gave at SEAS 7Oct1986 (European
SHARE on isle of jersey) loots of R3.0, R3.1, & R3.2 was after I had
graduated and joined scientific center and stuff I had done at the univ
as undergraduate in the 60s (one of my hobbies was enhanced production
systems for internal datacenters).
https://www.garlic.com/~lynn/hill0316g.pdfrepeated 16Mar2011 at Wash DC HILLGANG user group meeting.
In the morph from CP67->VM370, they dropped and/or simplified a lot of
stuff. 1974, I started migrating a bunch of stuff to VM370R2 base for my
internal CSC/VM ... including kernel reorganization for SMP
multiprocessor (but not the SMP support itself). Then for VM370R3-base
CSC/VM I do SMP support ... originally for the internal online
sales&marketing HONE systems (before it was released to customers in
VM370R4).
-- virtualization experience starting Jan1968, online at home since Mar1970