We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

AR# 24050

8.2i EDK SP2 plb_v34_v1_02_a - System hangs when the PLB master cannot access the PLB to complete a write operation due to a read request in progress


The plb_v34_v1_02_a arbiter does not allow simultaneous use of the PLB RdDBus and the WrDBus. The latest plb_v34_v1_03_a bridge corrects this system hang condition and also improves system performance in most systems by allowing simultaneous use of Rd and Wr data buses of the PLB.


An updated plbv34 arbiter is available in plb_v34_v1_03_a. Note that as with plb_v34_v1_02_a, the plb_v34_v1_03_a must be used with the plb2opb_bridge_v1_01_a bridge. The new arbiter allows for the simultaneous use of the PLB RdDBus and the WrDBus when system traffic calls for such operation.

In a PLB/PCI system, a system hang caused by the plb_v34_v1_02_a arbiter not allowing simultaneous use of the PLB RdDBus and the WrDBus occurs. Note that this does not occur with OPB/PCI systems. In the PLB/PCI system, a remote PCI device is read by a remote PLB master, but the remote PCI device follows PCI protocol, which allows it to complete write operations to the PLB before it completes the request for read data by the remote PLB master. However, the v1.02.a arbiter does not allow simultaneous access to the PLB WrDBus and the system hangs. With the new PLB arbiter, the remote PCI initiator can access the PLB write bus and complete the write, and subsequently provide read data to the remote PLB master and the system continues operation.

Please contact Technical Support for the core and patches.

AR# 24050
Date 12/15/2012
Status Active
Type General Article