Answers Database M1.4: How to setup and debug Multi-Pass Place and Route/Turns Engine/Networked PAR Record #1644 Product Family: Software Product Line: FPGA Implementation Product Part: par Product Version: 1.4. Problem Title: M1.4: How to setup and debug Multi-Pass Place and Route/Turns Engine/Networked PAR Problem Description: Urgency: Hot General Description: Multi-Pass Place and Route (MPPR) is a feature that allows you run multiple iterations of Place and Route (PAR) on a single machine (single node MPPR), or on multiple workstations (multi-node MPPR). This feature is useful for finding place and route solutions for designs that are close to meeting timespecs or are finishing with only a few unroutes. Solution 1: INPUT FILES:        o Single node MPPR requires only one input file: a       mapped <design>.ncd.        o Multi-node MPPR requires 2 input files: a mapped <design>.ncd,       and a nodelist file. A nodelist file is user-created ASCII file       that lists the workstation names. HOMOGENEOUS/HETEROGENEOUS NETWORKS:        o Multi-node MPPR can run on homogenous networks: all SunOS, all       Solaris, or all HP-UX.        o Multi-node MPPR can also run on heterogeneous networks: mix of SunOS,       Solaris, and HP-UX. The Xilinx software and license environment must       be set appropriately in the .cshrc file for each platform on which       you intend to run. See the sample .cshrc file below which sets the       environment differently based on the platform being used. This is       necessary because the remote nodes read their environment variables       from the .cshrc, they do not receive them from the launching node. LIMITATIONS:        o MPPR can only operate on Xilinx FPGA families: XC3000a, XC4000e,        XC4000ex, XC4000XL, XC5200 and Spartan. It can not operate on        CPLD's: XC7000, XC9500.        o Although single node MPPR can operate on any platform in the M1        release, multi-node MPPR can only operate on UNIX workstations.        o Each node uses a single license while running. The maximum number        of nodes that can operate in parallel is limited to the number of        licenses available.        o Each run targets the same part, and uses the same algorithms and        options. Only the starting point, or the cost table entry, is        varied. ENVIRONMENT VARIABLES:        o PAR_AUTOMNTPT - Multi-node MPPR uses network path names to access       files. By default, PAR builds a network path to specified input       files assuming an auto-mounted network using /net/<machine name> as       the mount point. For example, a local path name to a file may be       /usr/ivan/designs/cpu.ncd. PAR will by default build a network path       name of /net/<machine name>/usr/ivan/designs/cpu.ncd.       If the network is auto-mounted, but does not use /net/<machine name>       as the auto-mount point, the PAR_AUTOMNTPT variable can be used to       set the network auto-mount point. For a network that uses       /xyz/<machine name> as the auto-mount point, the PAR_AUTOMNTPT       variable should be set to "xyz".       For hard-mounted networks, the default network paths generated by PAR       will not be valid. In this case, PAR must be run from a location       where the directory path is valid on all remote machines being used.       The PAR_AUTOMNTPT variable must be set to "/". This tells PAR not to       build auto-mount paths on remote machines, but to use $PWD instead.       An alternative is to set PAR_AUTOMNTPT to "" and use full paths to       all files specified on the command line.        o PAR_AUTOMNTTMPPT - Most networks use /tmp_mnt as a temporary mount       point. If your network uses a temporary mount point by a different       name, like /t_mnt, then you must set the PAR_AUTOMNTTMPPT variable       to the temporary mount point name; in this example you would set       PAR_AUTOMNTTMPPT to /t_mnt. The default value for PAR_AUTOMNTTMPPT       is /tmp_mnt. This variable only needs to be set for auto-mounted       networks that do not use /tmp_mnt as the temporary mount point.        o Multi-node MPPR logs onto a node and then invokes par. The       environment variables on the node are read from the .cshrc (or       equivalent), they are not passed from the host to the node.       Therefore all Xilinx environement variables must be defined in the       .cshrc, otherwise the par process on the node will not be able to       find the software or the licenses. DOCUMENTATION:       o The online documenation discusses MPPR further. From the command line         type dtext, to start the dynatext browser. Make the following         selections: Development Systems Reference Guide-> Chapter 10 - PAR ->         Turns Engine (PAR Multi-Tasking Option). INVOCATION:       o Command Line =>                {The 2 lines following represent 1 command}                par -m <nodelist file> -n <# of iterations>         -s <# of iterations to save> <mapped desgin>.ncd <output         directory>.dir -m <nodelist file> => Use only for multi-node MPPR. Omit for single node MPPR. Use absolute local path name. -n <# of iterations> => Specifies number of Place and Route passes. -s <# of iterations to save> => Only the best -s results are saved. <mapped design>.ncd => Input ncd file. <output directory>.dir => Directory where the best -s results are saved. Files include placed and routed ncd, summary timing reports (.dly), pinout files (.pad), and log files (.par).       o Graphical User Interface =>        => Open the design manager: dsgnmgr.        => Open a project in the design manager.        => Select a version. MPPR will not be available if a           revision is selected.        => Select Implement->FPGA Multi-Pass Place and Route.        => In the pop up dialog, enter the number of iterations to run           and to save. The starting strategy determines the initial point           for the algorithms. As each PAR process is initiated, its           starting strategy is incremented by 1. Varying the starting           strategy produces varying routing and timing results; it is           impossible to predict which starting strategies will produce           the best results. It is a "Try and See" variable.           For single node MPPR, leave the nodelist field blank, for           multi-node MPPR enter the full path name to the nodelist file. DEBUG:       o With multi-node MPPR you may receive messages from the login process.         The problems are either network related environment variable related. Network: You may not be able to logon to the machines listed in the nodelist file.                  Try to ping the nodes - ping <machine name>. You                  should get a message "machine is alive". ping should                  also be in your path (UNIX cmd: which ping). If                  returns packet information, then you should confirm                  that ping was run with no options. Ping must return                  the message "machine is alive" for MPPR to work.                  Try to logon to the nodes - rsh <machine name>. You                  should be able to logon to the machine. If you can                  not, make sure rsh is in your path (UNIX cmd: which                  rsh). If rsh is in your path, but you still can not                  logon, contact your network administrator.                  Try launch par on node -                  rsh <machine name> /bin/sh -c par                  This is the same command that multi-node MPPR uses to                  launch par. If this command is successful, everything                  is setup correctly for <machine name> node.       Environment: Check the environment on the nodes - Logon to the                  problematic node (UNIX cmd: rsh <machine name>).                  Check the $XILINX, $LD_LIBRARY_PATH, and $path                  variable (UNIX cmd: echo $<variable name>). If these                  variables aren't set correctly, then check to make sure                  these variables are defined in your .cshrc file. Note,                  some, but not all, errors in reading the .cshrc may                  prevent the rest of the file from being read; these                  errors may need to be corrected before the XILINX                  environment variables in the .cshrc are read.                  /bin/sh: par not found - This message indicates that                  the environment in the .cshrc is not being correctly                  read by the node. SAMPLE FILES: -------------------------- .cshrc file segment ------------------------------- # This is a comment. # This setup enables using multi-node MPPR with homogeneous and heterogenous networks: # Solaris, SunOS, and HP-UX. When a login occurs, the bintype variable is # set to the node platform, which then points the Xilinx environment variables # to the platform specific software and libraries. if (!($?bintype)) then    set hosttype = `uname -s -r`    if ("$hosttype" =~ SunOS\ 4.1.[34]*) then      set bintype = sun    else if ("$hosttype" =~ SunOS\ 5.[45]*) then      set bintype = sol    else if ("$hosttype" =~ HP-UX\ [AB.0-9]*) then      set bintype = hp    else      echo "Unknown Hostype"      set bintype = unknown    endif endif setenv XILINX /products/xilinx/m1 setenv LD_LIBRARY_PATH ${XILINX}/bin/${bintype}:${LD_LIBRARY_PATH} setenv LM_LICENSE_FILE /poducts/license/combined.flexlm:/xilinx/flex/license.dat setenv XILINX_SERVER "license_server1;license_server2" set path=($path ${XILINX}/bin/${bintype}) NOTE: The .cshrc file will often contain a line that terminates execution early for non-interactive shells. A common example is the following:     if ($?USER == 0 || $?prompt == 0) exit If the Xilinx environment is not not set prior to this line, all multi-node MPPR jobs will fail because the remote shells will not have the proper environment. All stty commands should be placed after this line in the .cshrc file. Otherwise, stty error messages will be seen while running the Turns Engine: "stty: Invalid Argument". ------------------------- nodelist file below -------------------------------- # This is a comment # Note: machines are accessed by multi-node MPPR from top to bottom # Sparc 20 machines running Solaris (SunOS 5.5) kirk spock mccoy krusher janeway picard # Sparc 10 machines running SunOs (4.1.4) michael jermaine marlon tito jackie # HP's running HP-UX 9.05 william george ronald jimmy gerald End of Record #1644 - Last Modified: 02/17/98 08:30

For the latest news, design tips, and patch information on the Xilinx design environment, check out the Technical Tips!