Friday, November 6, 2009

IBM PowerPC™ 405-S: Verification of the RTL with the supplied Artisan SRAMs

[Please note that these instructions reference the 1.00a version of the Synopsys PowerPC™ 405-S coreKit which is the version provided in the University IP package from IBM. There are updated versions available from Synopsys that address several of the issues outlined below. These are specifically commented on where known.]

This is the second article in the series (click here for the series index) describing the IBM PowerPC™ 405 synthesizable core and peripherals and will cover what needs to be done to take the IP and get the RTL simulating with the modern versions of the necessary Synopsys Tools and a simulator of your choice.

Initial Unpacking
When you first unpack the DesignWare PowerPC405-S™ implementation view coreKit (the archive is called DW_ppc405-FS_iv_1.0a_AV.tar.gz) it expands to a directory DW_ppc405-FS_iv_1.0a_AV.

The first thing you should do is begin reading the excellently written User Guide (docs/405_ivug.pdf) which provides the suggested flows along with the Release Note (docs/iv_rn.pdf) that specifies the tools and environment with which you need to build the processor.

However the software versions requested by the release note to simulate the RTL is problematic. I have listed below the required software in black, the lowest version available in red and the highest version available at time of writing in blue:

Required:
  1. Synopsys coreConsultant, Version V-2004.03-CT4.4.2 Z-2007.06-SP5 C-2009.06-3
  2. Synopsys Vera System Verifier, Version 6.3.0 Z-2006.12 A-2007.12-12
  3. Synopsys Design Compiler, Version U-2004.06-SP2 Z-2007.03-SP2 C-2009.06-SP2
One of:
  1. Synopsys VCS, Version 7.1 Y-2006.06-SP1 C-2009.06-1
  2. Mentor Graphics ModelSim, Version 5.7g 6.5c
  3. Cadence NC-Verilog, Version 5.1 8.1
As you can see the specified tool versions are simply not available.
[Note: at the time of publication the current commercial coreKit is version 3, not the version 1.0a discussed here and supports the current generation of tools.]

In the following sections I will describe how to get the PowerPC 405 verified with the following tool set:
  1. coreTools C-2009.06
  2. DC Ultra B-2008.09
  3. DesignWare 2009.06a
  4. vera A-2007.12
  5. NC-Verilog (Integrated Unified Simulator) 8.1
Before you proceed further be sure that all the necessary environment variables from the documentation are set and the correct executables are the in the path.

Getting Started

The educational distribution doesn't come packaged as a single coreKit file but as a compressed, already installed, coreKit directory. To get started fire up coreConsultant and select File > New Workspace. Point the installed coreKit path at the unpacked .tar.gz and choose a directory as the root and a new name for the Workspace. In the following examples we will use the workspace RTLVerification created in the same directory as the unpacked coreKit.

You should now have a directory tree in this root beginning with the name of the workspace. It is similar to the unpacked .tar.gz and populated with symlinks to some of the original files.

Choose the built in Artisan 130LVFSG RAM models and proceed to the simulation stage (I will discuss integrating custom SRAMs in a later article).

Verification by Simulation: Initial Experiences

When you have successfully moved onto the "Verify Component" stage things are not so simple. As we are going to use NC Verilog be sure to choose NC_Verilog as the simulator and ensure the relevant paths are set correctly. Leave the rest of the options at their default values.

[Note: All the issues listed here (and some more of the same) are solved with one comprehensive script and a few instructions in the section below. This section is here to detail what kind of issues you may face and how to solve them to allow you to debug your own installations should this be necessary.]

Building the Vera Objects

The design of the simulation/test environment uses Vera testbench tool to provide compiled simulation objects that surround the Verilog core and a few other Verilog modules and test vectors. We need to compile these Vera objects and we do that by opening a terminal in the Workspace directory we have created (RTLVerification). Then execute this series of commands in the terminal:

cd sim/vera/src

./build_vro


The script will generate several screens full of output with no errors (a few warnings do appear). You can check the content of the sim/vera/lib directory and see the compiled .vro objects.

Running the Simulation

When you hit "simulate" you are faced with the following error message:

For some reason several script files are not correctly linked into the RTLVerification/sim/ directory, but it is easy to fix.

Open the terminal and navigate into the sim/ directory:

cd sim/
ln -s ../../DW_ppc405-FS_iv_1.0a_AV/sim/build_pli_* ./
[Note that you must have the directories setup as discussed or change the command to where the you unpacked the original coreKit directory]

Unfortunately we are not quite out of the woods yet, as the build_pli_nc.sh has to be run by hand to build the binary link between Vera and NC_Verilog. If your environment variables are set correctly then this will complete without error leaving behind the binary file libpli.so.

[Note that you need to have a 64 bit version of Vera for a 64 bit simulator and 32 bit for 32 bit. If you are building for a 32 bit simulator and 32 bit Vera on 64 bit linux edit the pli script to add an -m32 immediately after "gcc" in the compile line and "gcc" in the link line, e.g.:

elif [ `uname` = "Linux" ]; then
     gcc -m32 -fPIC -c -I$VERA_HOME/lib/vlog -I$CDS_INST_DIR/tools/verilog/include $VERA_HOME/lib/vlog/veriuser.c
     gcc -m32 -shared -o libpli.so veriuser.o $VERA_HOME/lib/vlog/libSysSciTaskpic.a
]

Retrying the simulation in the coreConsultant GUI produces the following result almost instantly:

Now I think NC Verilog is an excellent simulator however simulating an entire core in a few seconds? No.

Delving into the scripts reveals another missing file: sim/runtest. By using the same procedure as before we can link it in:

cd sim/
ln -s ../../DW_ppc405-FS_iv_1.0a_AV/sim/runtest ./

Unfortunately this script is also not in a working condition: It depends on the perl distributed by Synopsys but it is hardcoded to try one only available within IBM. However fortunately Synopsys ships this perl with coreTools. Open the file sim/runtest in your favorite text editor and change the first few lines from:

#!/afs/raleigh/cadtools/synopsys/dc/2005.09/linux/bin/snpsperl
require 5.004;
BEGIN { @INC = ( "/afs/raleigh/cadtools/synopsys/dc/2005.09/linux/lib" ); }

to:

#!$YOUR_PATH_TO_CORETOOLS/linux/dware/bin/snpsperl

require 5.004;

BEGIN { @INC = ( "$YOUR_PATH_TO_CORETOOLS/linux/dware/lib" ); }

replacing "$YOUR_PATH_TO_CORETOOLS" with the absolute path in your installation.

Trying to simulate again produces the "instant success" we will come to know and love.

Another missing file that needs to be linked into our workspace is the Makefile used to run these simulations:

cd sim/
ln -s ../../DW_ppc405-FS_iv_1.0a_AV/sim/Makefile ./

Unfortunately this Makefile has a similar problem with hardcoded paths that need correcting, as does Report.pl, runTB and hexFormal.pl. There are also numerous files that are not linked in. These issues arise from the coreKit being delivered in an installed from compressed into an archive rather than as a .coreKit file. To save you the effort of solving every one of these issues have a look at the next section. :

Verification by Simulation: Solved
[Ensure you have followed the instructions in "Getting Started" above. You should have installed a new workspace and have completed "Specify Configuration" and moved onto "Verify the Component"]

1) Open your favorite text editor and paste in the following script:

#!/bin/bash
#
# fix_sim_dir.sh
#
#Copyright Matthew Swabey (matthew@swabey.org).
#
#You are licensed to use this file in conjunction with the IBM PowerPC 405 coreKit and to make modifications as needed to this script as long as this copyright notice is kept.

#This script should be executed from the the new workspace directory after you have installed it with coreConsultant and progressed past specifying the SRAM technology.

UNPACKED_COREKIT_DIR=/export/fastcad/disktmp/mas/IBM_PUB2/DW_ppc405-FS_iv_1.0a_AV
SNPS_PERL_EXE=/home/esdcad/software/synopsys/linux/coreTools_B-2008.12-SP1/linux/dware/bin/snpsperl
SNPS_PERL_LIB=/home/esdcad/software/synopsys/linux/coreTools_B-2008.12-SP1/linux/dware/lib

#Build the vera objects with the version of vera installed
cd ./sim/vera/src
./build_vro
cd ../../..

#Link in the files needed to build the simulation of the PPC and run the testbenches
ln -s $UNPACKED_COREKIT_DIR/sim/build_pli_* ./sim/
ln -s $UNPACKED_COREKIT_DIR/sim/Makefile ./sim/
ln -s $UNPACKED_COREKIT_DIR/sim/runtest ./sim/
ln -s $UNPACKED_COREKIT_DIR/sim/Report.pl ./sim/
ln -s $UNPACKED_COREKIT_DIR/sim/scripts/hexFormat.pl ./sim/scripts/
ln -s $UNPACKED_COREKIT_DIR/sim/scripts/run405.config ./sim/scripts/
ln -s $UNPACKED_COREKIT_DIR/sim/scripts/runTB ./sim/scripts/
mkdir ./sim/testsuite
ln -s $UNPACKED_COREKIT_DIR/sim/testsuite/* ./sim/testsuite/
ln -s $UNPACKED_COREKIT_DIR/sim/testbench/*.v ./sim/testbench/
ln -s $UNPACKED_COREKIT_DIR/sim/testbench/*.inc ./sim/testbench/
ln -s $UNPACKED_COREKIT_DIR/src/rtl/*.v ./src/rtl/

#script files needing corrected perl paths separated by a space.
TO_BE_FIXED='./sim/Makefile /sim/runtest ./sim/scripts/hexFormat.pl ./sim/scripts/runTB ./sim/Report.pl'

function fix_perl_paths {
#Fix the afs paths for snpsperl
sed -i "s:/afs/raleigh/cadtools/synopsys/dc/2005.09/linux/bin/snpsperl:${SNPS_PERL_EXE}:g" $1
sed -i "s:/afs/raleigh/cadtools/synopsys/dc/2005.09/linux/lib:${SNPS_PERL_LIB}:g" $1
#Fix added for runTB
sed -i "s:/global/freeware/SunOS/5.8/perl-5.6.1/bin/perl:${SNPS_PERL_EXE}:g" $1

}

for file in $TO_BE_FIXED; do fix_perl_paths "$UNPACKED_COREKIT_DIR/$file"; done

#End of script

2) Edit the top three variables, UNPACKED_COREKIT_DIR, SNPS_PERL_EXE & SNPS_PERL_LIB to have the absolute paths to the appropriate executable or directory. There should be no spaces in the paths.


3) Save this file with a suitable name (e.g. fix_sim_dir.sh) in the new workspace you have created. Make the script executable using the command chmod u+x "filename". When you run it you should see an output that looks like this (but with many more Vera banners):

++---------------------------------------------------------------------++
 ||                     VERA System Verifier (TM)                       ||
 ||        Version: A-2007.12 () -- Thu Oct  8 11:35:39 2009            ||
 ||             Copyright (c) 1995-2004 by Synopsys, Inc.               ||
 ||                      All Rights Reserved                            ||
 ||                                                                     ||
 ||      For support, send email to vera-support@synopsys.com           ||
 ||                                                                     ||
 ||  This software and the associated documentation are confidential    ||
 ||  and proprietary to Synopsys Inc.  Your use or disclosure of this   ||
 ||  software is subject to the terms and conditions of a written       ||
 ||  license agreement between you, or your company, and Synopsys, Inc. ||
 ++---------------------------------------------------------------------++
Warning: async drive on skewed signal jtag_o.tdi, at 78 in ./p405s_jtag.vr
Warning: async drive on skewed signal jtag_o.tms, at 79 in ./p405s_jtag.vr
Compilation errors: 0
 ++---------------------------------------------------------------------++
 ||                     VERA System Verifier (TM)                       ||
 ||        Version: A-2007.12 () -- Thu Oct  8 11:35:39 2009            ||
 ||             Copyright (c) 1995-2004 by Synopsys, Inc.               ||
 ||                      All Rights Reserved                            ||
 ||                                                                     ||
 ||      For support, send email to vera-support@synopsys.com           ||
 ||                                                                     ||
 ||  This software and the associated documentation are confidential    ||
 ||  and proprietary to Synopsys Inc.  Your use or disclosure of this   ||
 ||  software is subject to the terms and conditions of a written       ||
 ||  license agreement between you, or your company, and Synopsys, Inc. ||
 ++---------------------------------------------------------------------++
Compilation errors: 0
 ++---------------------------------------------------------------------++
 ||                     VERA System Verifier (TM)                       ||
 ||        Version: A-2007.12 () -- Thu Oct  8 11:35:39 2009            ||
 ||             Copyright (c) 1995-2004 by Synopsys, Inc.               ||
 ||                      All Rights Reserved                            ||
 ||                                                                     ||
 ||      For support, send email to vera-support@synopsys.com           ||
 ||                                                                     ||
 ||  This software and the associated documentation are confidential    ||
 ||  and proprietary to Synopsys Inc.  Your use or disclosure of this   ||
 ||  software is subject to the terms and conditions of a written       ||
 ||  license agreement between you, or your company, and Synopsys, Inc. ||
 ++---------------------------------------------------------------------++
Compilation errors: 0
ln: `./sim/testbench//p405s_rambist_tb.v': File exists
ln: `./sim/testbench//tb_p405s_dataram_64X34_wrapper.v': File exists
ln: `./sim/testbench//tb_p405s_sram256x46_wrapper.v': File exists
ln: `./sim/testbench//tb_p405s_sram256x48_wrapper.v': File exists
ln: `./sim/testbench//tb_p405s_sram512x128_wrapper.v': File exists
ln: `./sim/testbench//tb_p405s_sram512x32_wrapper.v': File exists
ln: `./sim/testbench//tb_p405s_sram512x8_wrapper.v': File exists
ln: `./sim/testbench//tb_p405s_sramBytWr512x128_wrapper.v': File exists
ln: `./src/rtl//PPC405F5V1_soft.v': File exists

At this point the script will have built the vro objects using Vera, linked in all the missing files and corrected the paths in the scripts for you.

4) Enter the ./sim directory in the workspace and run the build_pli_nc.sh script by hand to build the binary link between Vera and NC_Verilog. If the environment variables are set correctly then this will complete without problem leaving behind the binary file libpli.so
[Note that you need to have a 64 bit version of Vera for a 64 bit simulator and 32 bit for 32 bit. If you are building for a 32 bit simulator and Vera on a 64 bit computer edit the scripts to add an -m32 immediately after "gcc" in the compile line and "gcc" in the link line:

elif [ `uname` = "Linux" ]; then
     gcc -m32 -fPIC -c -I$VERA_HOME/lib/vlog -I$CDS_INST_DIR/tools/verilog/include $VERA_HOME/lib/vlog/veriuser.c
     gcc -m32 -shared -o libpli.so veriuser.o $VERA_HOME/lib/vlog/libSysSciTaskpic.a
]

5) Return to the coreConsultant GUI, check you have selected the correct test suite, simulator and other variables and click "Apply" to be rewarded with:



This will take quite a while, so enjoy a well earned cup of tea/coffee/other until it completes:


6) Check the reports for anything interesting or anomalous

7) ...

8) Profit!

Published with permission from IBM and Synopsys

IBM PowerPC™ 405-S: Getting started

Synthesizable RTL versions of IBM's excellent embeddable PowerPC™ line of processors (the 405, 460 and the 460 FPU) are available to SoC designers from Synopsys' Star IP Program and this Star IP PowerPC™ 405-S is also available to Educators. Further embeddable PowerPC™ cores for synthesis are also available from AMCC.

In a series of articles I will show how to take the educational soft IP download, verify the 405 core and put it to work bonded to some of Synopsys' DesignWare IP.

Currently the articles in this series are:

1) Getting started with the synthesizable IBM PowerPC405™ (this article)
2) Verification of the RTL with the supplied Artisan PDK (now owned by ARM)
3) Synthesis of the RTL with the supplied Artisan PDK
4) Verification of the Gate Level Netlist


It was with great excitement that I received the educational IP by electronic transfer. As is typical with this kind of IP the full details remain secret until the licenses are signed, but with the kind permission of IBM and Synopsys let me lift the veil a little bit for you:

Here is what you get:
  1. Synopsys PowerPC™ 405-S Design View CoreKit v1.00a - This enables full, accurate simulations to be carried out to allow evaluation of the core running a software stack
  2. Synopsys PowerPC™ 405-S Implementation View CoreKit v1.00a - This bundle contains the verilog RTL of the actual core itself along with a demonstration technology library and the verification tools to enable its functionality to be fully exercised and guaranteed. This view also contains implementation scripts to build the design from RTL through synthesis, DFT, floorplanning, placement and clock tree expansion.
  3. Specifications - An essential set of manuals describing the exact operation of the core
  4. DDR2 Memory Controller - IBM's PLB DDR2 SDRAM memory controller to allow attachment of off chip SDRAM
  5. SRAM Memory Controller - This allows a number of internal SRAM (not SDRAM!) blocks to be presented as one continuous memory space on the PLB bus to a master
  6. PCIX Controller - IBM's PLB to PCIX controller
  7. PLB Crossbar Arbiter - This enables the connection of a large number of PLB masters to one or more slave busses. With this block very complex multicore/multimaster bus structures can be created within an SoC
  8. DMA Controller - a standard component allowing DMA transfers between peripherals and memory and vice versa
  9. UIC Controller - This is the Universal Interrupt Controller, an essential device for any SoC 
  10. PLB to AHB Bridge - This is bundled in with the core but is well worth noting - this allows the use of DesignWare IP blocks which are based around the AMBA bus specification
  11. Artisan PDK - For simulation purposes an Artisan 130nm LV PDK is supplied
  12. IBM RISCWatch - A JTAG debugger to allow full trace and debugging functions on the PowerPC core
As you can see it is an embarrassment of riches - a near complete set of IP enabling a basic single core to a complex multicore PowerPC™ 405 based SoC to be realized with only the basic IO IPs outstanding (e.g. no serial port or GPIO).

Again I should like to say a heartfelt thankyou to IBM and Synopsys for making this IP available to Universities to drive our research and educational programs forward. It is also available to drive commercial embedded designs forward after licensing.

Note: As it is a Synopsys Star IP this distribution is designed for a number of industry standard simulators and the Synopsys Implementation and Signoff tool suite. The package contains pure Verilog RTL so it is technically possible to use another vendor's synthesis tool but that is well outside the scope of this series of articles (as well as being a very big job).

Published with permission from IBM and Synopsys

Wednesday, November 4, 2009

How to Re-Analyze the DesignWare Building Blocks in Synopsys Design Compiler

Design Compiler, sometimes misnamed as dc_shell, includes a selection of datapath primitives called DesignWare Building Blocks (DWBB). Sometimes after a patch or new installation you may find these complaints in your logs:

Error:   Source file for 'DW01_SUB' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)
Error:   Source file for 'DW01_DEC' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)
Error:   Source file for 'DW01_INC' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)
Error:   Source file for 'DW01_ADD' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)
Error:   Source file for 'DW01_INC' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)
Error:   Source file for 'DW01_INC' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)
Error:   Source file for 'DW01_ADD' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)
Error:   Source file for 'DW01_INC' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)
Error:   Source file for 'DW01_INC' was not analyzed by this release of the compiler; re-analyze it. (ELAB-343)

They mean the compiled DWBB library is now out of date in relation to the version of Design Compiler you have. To cure them cd to $SYNOPSYS, then

cd dw/scripts/
./dw_analyze_syn.csh

and pretty soon you will have a lovely, shiny, new set of DesignWare Building Blocks!

[Note that this information also applied to VCS but use
cd dw/scripts/
./dw_analyze_sim.csh
]

Friday, September 25, 2009

"Transmute" - Design Philosopies

I realized that I had forgotten to describe a very, very important step: Philosophy. In this case I am using the word incorrectly. It is actually World View that I mean. From Wikipedia:
"A comprehensive world view (or worldview) is the fundamental cognitive orientation of an individual or society encompassing natural philosophy, fundamental existential and normative postulates or themes, values, emotions, and ethics."
Now what has this to do with chip design? Well you might ask! In this case it is the fundamental underpinnings and assumptions that surround the design.

The world view is formed from facts and opinions and our emotional reaction to them.

So here is the list of suitable facts for the "Transmute" project:
  • We have only built cell based custom digital cores
  • None of our typical projects are as large as this
  • The design flows are not stable - considerable development will be required
  • Several custom digital modules will be required to interfaced to licensed IP
  • We can experiment with power saving strategies
  • We are very experienced with custom analogue blocks
What, as a designer, is my emotional reaction to this? Simply this: Excitement and Concern. There are a lot of unknowns.

Hence our design World View (although a different design will have a different world view):

  • Digitally Conservative
  • Analogue Progressive
One you articulate this World View clearly you realize that this is the guide for your choices, in our case:
  • Hard IP: Eliminates the risk from core synthesis
  • Digital part to be separable from the analogue part so we can drive them separately. This will minimize the cost in case of failure or error - we would get at least one working part from the chip
  • Digital power saving strategies to be applied only after a high confidence in non-power saving digital structures obtained
  • It must be possible to disable the digital power saving structures if they are added
  • Flow decisions should favour tools which we have experience with
These decisions then guide the rest of the design flow. Also new choices can be matched against the worldview ensuring consistency.

Just bear in mind, your world view for this design may need modifying as the project progresses however if it does then you need to reconsider most of the work you have already carried out. This is typically far more important if your original plan was aggressive and you move to conservative - check work and assumptions already made!

Wednesday, September 16, 2009

The next SoC design - working title "Transmute" begins

I have a fun new hobby - designing the digital part of a new, mixed signal SoC which we are building here. It is my intention to blog the full design flow discussing the various challenges and issues as we encounter them.

Design Basics
  • Foundry
  • Technology
  • IP
  • Flow
The importance of these choices of these depends on the following thoughts:
  • What have we done before
  • Who have we worked with before
  • What have we got that works
  • What we need to do and what we would like to do
Planning a chip begins after you decide the basics above, i.e. foundry, technology and the needed parts and finally tool support. Sometimes in industry you have the luxury of a couple of other choices e.g.:
  • Cell library (we use the ST supplied one but there are others, Faraday for example)
Basic Overview Spec
  1. Two full custom ADCs for testing and evaluation
  2. A single core SoC to test our mixed signal integration and apply tests
  3. Simple connectivity to the SoC
Initial Decisions

In our case after some mulling over our experiences we decided on:
  1. ST Microelectronics via the Circuits Multi-Projets MPW broker
  2. Their HCMOS9 130nm process
  3. IP Blocks (see below)
  4. The flow, more on that later
 IP Blocks

As a university our access to IP is somewhat limited. Making a SoC requires quite a few modules like memory controllers etc.

However to the rescue rides Synopsys' DesignWare Library. It includes a basic selection of AMBA AHB/APB bus connected peripheral set and a memory controller which supports SDRAM and Flash.

To drive the the DesignWare library is the coreTools GUI (which is surprisingly hard to find) which in a nice graphical environment enables you to construct AMBA bus structures containing DesignWare peripherals. So this drives our choice of some of the tools (coreTools and DesignCompiler Ultra) as well as their DesignWare basic IP.

Onto a much more interesting question. What core to use? Considering our other IP is AMBA an ARM is the most likely choice however there are others we might use:
  • The LEON3 from Aeroflex is a Opensource (GPL), VHDL 32 bit SPARCv8 CPU with an AMBA bus. It is, however, very much bound to its peripheral library and autoconfiguration so would take a lot of work. It would also need to be customised to use the HCMOS9 SRAM blocks
  • The IBM PPC 405 is a 32 bit PowerPC core designed for embedding which is licensed to universities for teaching and education. It is the distribution created for the Synopsys flow which allows easy integration with AMBA peripherals. It is a softcore, i.e. it would need to be synthesised from RTL
In this case we wish to reduce our risk and design effort. Licensable from CMP is a foundry guaranteed ARM946E-S hard IP (i.e. they have already synthesized it and checked it) on their 130nm process. So we are going ahead with DesignWare soft IP surrounding a hard IP ARM946E-S. The benefits of hard IP are that ST have de-risked the design for us.

The features of the ARM946E-S are:
  • Excellent compiler/OS/application support. ARM is one of the best supported architectures in both Opensource and commercial software
  • Small and quick, approximately 3mm2 and clocking at 200MHz (both of these are significantly derated estimates at this stage of the design
  • MMU allowing full OS support
In addition we need the following types of IP:
  • A PLL - this is pretty much required for any SoC these days (available from CMP)
  • Level shifters and isolators - needed to try advanced power saving strategies in digital logic
The current initial design of the digital segment of our chip is best shown in a block diagram below:




We will talk of flows and the analogue integration later.....

Thursday, September 3, 2009

The sheer size of an old Sun "midrange" server

This one is going back a bit but I have obtained it for legacy support. Its a Sun V880 with 6x UltraSPARC III 750MHz processors and 12GB of RAM. Still a computer to be reckoned with [Update: Spent $50 on a new CPU/RAM board to take it to 8x CPU and 16GB RAM. eBay is great for this kind of thing]



Its as high as my desk, as large as a fridge lying on its side and sounds like a very loud aircon. [Update: When it is running under an OS at low load it throttles the fans and becomes quite quiet]

This one was going to be built with Solaris 10 on a software mirrored root using UFS but the new spin of Solaris 10 allows installation and booting off ZFS. A much more sensible choice when I have 6x 36GB disks.

Thursday, August 20, 2009

Modding an HP Proliant DL585 to take an EVE Zebu UF2

I am the proud "controller" of an EVE Zebu UF2 emulation accelerator (a big PCI card covered in FPGAs). One of its biggest selling points is their Transactors which allow you to present your design with the interfaces it would experience if it were real, completely abstracting away the PCI bus.

However when I say big I mean triple height and full length. It uses 64 bit PCI/66. (For those considering purchasing a UF4 card most PCs can't fit it as it is larger still).

So I need a computer to host it in!

Enter the HP DL585 - a lovely machine. Built as solidly as a tank and everything is terribly easily accessible either from the front or the top. Everything. At a push I think I could swap out the motherboard and be back up and running in 30 minutes from cracking the case.

The original engineers designed it for cable-less operation (and very nearly succeeded!) but the EVE ZeBu-UF needs power from a 4 pin molex (found on most pre-SATA HDDs).

So what to do? Spotting this SCA backplane and knowing that HDDs take 12V, 5V and 3.3V power I had my solution. Mod a molex connector onto the pins on the back of one of the SCA sockets.

Firstly I checked the specifications for the ZeBu: 25W at 12V which is a current of 2.08A. Secondly I checked the specifications of the SCA2 connector: Section 6.4.1 of the SFF-8046 standard specifies a surge current of 5A and a constant supply of 2.5A. This means that one socket can supply the necessary power.

However, several provisos first: This will void your warranty. The SCA-2 connector transfers power to the drive using several pins for each voltage (12V, 5V and 3.3V) to reduce their resistivity to high current. By design the clusters of pins supplying the same voltage are next to each other making it easier. Probe for the resistance between the pins you think are the right ones - it needs to be zero.

The mod requires soldering the little clusters of pins together with fine wire (get some strands out of a multistrand wire) and then attached the cables. Note - don't include the Precharge pins - wiring them to the rest of the supply could cause unexpected operation as they will no longer function correctly when you insert a HDD in the front. After you are done, check, double check and triple check. Use a magnifier/microscope to check the soldering. Clean the area of any bits of solder. Probe the resistivity again. This could be an expensive mistake if there is a problem!

Power on (fingers crossed) and bingo - 12V and 5V.

Here is a photo of the modded card in position:


...and a photo of the UF2 in the PCI bay (I have two power connectors just in case I can get another UF2 - wishful thinking I know):

Laird Tpcm 7250 is as good as Honeywell PTM7950 as thermal paste / interface for PC

[This is not very scientific, however it is notable. At 7.5W/m-K vs the installed SYY-157 at 15.7 W/m-K it performed better in real world lo...