EasyManua.ls Logo

AMD AMD5K86 - Cache

AMD AMD5K86
416 pages
Print Icon
To Next Page IconTo Next Page
To Next Page IconTo Next Page
To Previous Page IconTo Previous Page
To Previous Page IconTo Previous Page
Loading...
AMD~
AMD5J1l6
Processor
Technical
Reference
Manual
18524BjO-Mar1996
6.2
Cache
The
cache
ability
of
SMM
memory
has
both
advantages
and
dis-
advantages.
By
caching
SMM
memory,
the
advantage
of
faster
repetitive
accesses
is
offset
by
delays
due
to
overwriting
cache
lines
that
may
otherwise
be
reusable
after
returning
from
SMM.
If
the
program
that
was
running
prior
to
entry
into
SMM
ran
out
of
the
cache,
and
the
same
program
continued
to
run
after
the
return
from
SMM,
the
processor
would
need
to
refill
the
caches
with
the
same
information
after
returning
from
SMM.
If
an
SMM
routine
frequently
accesses
the
same
loca-
tions,
the
delays
due
to
cache
refills
and
writeback-invalidates
may
be
worthwhile.
But
if
an
SMM
routine
seldom
accesses
the
same
locations,
the
speed
of
returning
and
continuing
on
with
the
prior
program
might
be
improved
by
not
caching
SMM
memory.
If
SMM
memory
space
overlaps
main
memory
space
that
is
cacheable,
PLUsH
must
be
asserted
when
Sl\I[
is
asserted
so
that
memory
accesses
in
SMM do
not
hit
locations
cached
from
main
memory.
If
SMM
memory
is
to
be
cacheable,
FLUsH
must
also
be
asserted
with
Sl\I[
when
entering
SMM,
and
the
SMM
service
routine
must
execute
the
WBINVD
instruction
to
invalidate
the
caches
just
prior
to
executing
the
RSM
instruc-
tion,
which
returns
the
processor
from
SMM.
The
use
of
FLUsH
or
WBINVD
adds
potentially
significant
time
to
the
entering
and
leaving
of SMM.
Systems
with
multiple
bus
masters
that
share
cacheable
mem-
ory
require
methods
for
controlling
access
to
the
bus
and
con-
trolling
the
coherency
of
shared
memory.
The
sections
below
summarize
certain
principles
and
methods
used
by
system
logic,
in
concert
with
software,
to
maintain
the
coherency
of
the
processor's
level-l
(or
Ll)
on-chip
caches
and
optional
level-2
(or
L2)
external
cache.
The
internal
architecture
of
the
processor's
Ll
instruction
and
data
caches
is
described
in
Section
2.3
on
page
2-13.
The
oper-
ating
system
writes
the
cache
disable
(CD)
and
not-
writethrough
(NW)
bits
in
CRO
to
enable
and
disable
caching,
independent
of
hardware.
Thereafter,
the
operating
system
may
write
the
PCD
and
PWT
bits
in
the
page
directory
and
System
Design

Table of Contents

Related product manuals