RAM 6.0 BUX II Series Betriebsanweisung Seite 16

  • Herunterladen
  • Zu meinen Handbüchern hinzufügen
  • Drucken
  • Seite
    / 36
  • Inhaltsverzeichnis
  • LESEZEICHEN
  • Bewertet. / 5. Basierend auf Kundenbewertungen
Seitenansicht 15
VxWorks
Hardware Considerations Guide, 6.0
12
most applications, this is not a problem. However, for applications with extremely
low latency requirements, this can be ameliorated somewhat by locating device
buffers in VME memory.
Read-Modify-Write (RMW)
Read-modify-write (RMW) must be provided in an indivisible manner.
The board must adhere to the RMW mechanism defined in the VME specification;
namely, a master must keep the address strobe low between the read and the write
portions of the cycle. A correctly functioning dual-ported slave board keeps the
RMW indivisible across both ports by detecting an address strobe that remains
low.
Unfortunately, keeping the address strobe low is only viable if you are reading and
writing the same single address. A more complicated indivisible instruction, such
as CAS, that involves multiple addresses cannot use this, and thus has no correct
mechanism for dual-ported memory. Because VxWorks uses only TAS, this is not
an issue. Some vendors have added a LOCK pin to the P2 bus for this reason.
However, the pin is not part of the standard and is therefore insufficient support
for this mechanism. For most boards, this is not an issue.
Caching and/or snooping can be an issue for VME RMW cycles. The shared
memory master board must not introduce any cache coherency problems. It must
be non-cached, or protected by full snooping capabilities, in order to handle proper
VME slave accesses.
For some PowerPC implementations, it has been necessary to use bus arbitration
as a global semaphore for VME RMW operations. When a board cannot generate,
nor respond, to RMW cycles, using the bus as a global semaphore works. Any
board that cannot use RMW, arbitrates for and holds the VME bus while a read and
write cycle is completed. In addition, the bus master board, where the shared
objects are stored, must implement the same bus lockup protection, even if the
master board can do RMW cycles correctly. This scheme is implemented in the BSP
sysBusTas( ) and sysBusTasClear( ) functions.
Arbitration
The board should default to bus request level 3 and provide a jumper mechanism
if alternative arbitration levels are supported.
It is often convenient to be able to select the manner of bus-release that can be RWD
(release when done), ROR (release on request), or RAT (release after timeout).
Seitenansicht 15
1 2 ... 11 12 13 14 15 16 17 18 19 20 21 ... 35 36

Kommentare zu diesen Handbüchern

Keine Kommentare