summaryrefslogtreecommitdiffstats
path: root/bsps/powerpc/mvme5500/README.VME
diff options
context:
space:
mode:
Diffstat (limited to 'bsps/powerpc/mvme5500/README.VME')
-rw-r--r--bsps/powerpc/mvme5500/README.VME19
1 files changed, 19 insertions, 0 deletions
diff --git a/bsps/powerpc/mvme5500/README.VME b/bsps/powerpc/mvme5500/README.VME
new file mode 100644
index 0000000000..f669f94d1b
--- /dev/null
+++ b/bsps/powerpc/mvme5500/README.VME
@@ -0,0 +1,19 @@
+README.VME: written by S. Kate Feng <feng1@bnl.gov> , 7/22/04
+
+
+Some VME modules(e.g. Oms58 motor controller) might require a PCI sync
+command following the out_xx() function (e.g. out_be16()) if mvme5500 is
+used as the SBC. The mechanism is a hardware hook to help software
+synchronize between the CPU and PCI activities. The PCI sync is
+implemented in pci/pci_interface.c. For more example of the usage,one
+can reference the drvOMS58.cc file that is posted in synAppRTEMS of
+http://www.nsls.bnl.gov/organization/UserScience/Detectors/Software/Default.htm.
+
+
+In spite of the PCI sync overhead for the Oms58 motor controller, I do
+not see the runtime performance of RTEMS-mvme5500 being compromised as
+compared with that of RTEMS-mvme2307. For example, it takes the same
+time to run motor_init() of synAppRTEMS for 48 motor initializations
+running either RTEMS-mvme2307 or RTEMS-mvme5500.
+
+