EasyManua.ls Logo

AMD AMD5K86 - Paging and the Tlbs

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~
18524B/O-Marl996
AMD5J!36
Processor
Technical
Reference
Manual
2.4.2
2.4.3
occur
out
of
program
order
with
respect
to
writes
on
the
bus
to
memory.
In
a
strongly
ordered
memory-write
model,
writes
to
cache
and
memory
always
appear
in
program
order.
In
a
weakly
ordered
memory-write
model,
writes
to
cache
and
memory
can
occur
out
of
program
order
(that
is, a
write
to
cache
can
occur
before
a
prior
write
to
memory).
Weakly
ordered
systems
may
per-
form
better,
but
they
can
cause
problems
in
systems
with
mul-
tiple-caching
masters.
For
example,
errors
may
occur
in
weakly
ordered
systems
when
a
master
that
is
held
off
the
bus
contin-
ues
writing
to
exclusive
or
modified
lines
in
its
internal
data
cache
while
another
master
writes
to
memory.
Nevertheless,
the
strongly
ordered
AMD5
K
86
processor
supports
high
perfor-
mance
without
using
weakly
ordered
memory
writes
by
buffer-
ing
speculative
stores
in
the
store
buffer.
Read/Write
Reordering
The
processor
reorders
certain
types
of
cacheable
read
cycles
on
the
bus
ahead
of
certain
types
of
write
cycles.
Specifically,
any
read
that
hits
in
the
instruction
or
data
cache
is
promoted
ahead
of
a
write
in
the
store
buffer
if
the
read
is
not
from
the
same
location
to
which
a
write
in
the
store
buffer
is
to
be
writ-
ten.
The
reordering
allows
reads,
which
dominate
the
proces-
sor's
use
of
the
bus
in
Write
back
mode,
to
take
precedence
over
data
writes,
which
normally
occur
infrequently.
The
EWBE
signal
has
no
effect
on
this
reordering
of
bus
cycles.
Segmentation
The
instruction
cache
contains
a
copy
of
certain
fields
in
the
current
code-segment
descriptor.
The
information
is
used
dur-
ing
prefetch
for
segment
translation
(logical-to-linear
addresses),
thus
providing
linear-address
tags
for
the
instruc-
tion-cache
entries.
Likewise,
the
load/store
units
hold
the
cur-
rent
data-segment
descriptors,
which
are
used
to
generate
the
linear
address
and
perform
protection
checks
during
data-
cache
accesses.
The
processor
can
cache
segment
descriptors
in
its
data
cache.
Memory
Management
Unit
(MMU)
2-27

Table of Contents

Related product manuals