From arthur.marsh at internode.on.net Wed Jan 7 01:14:34 2009 From: arthur.marsh at internode.on.net (Arthur Marsh) Date: Tue Jan 6 21:18:32 2009 Subject: [Open-graphics] any more news? Message-ID: Hi, is there any news on the development/implementation front? Arthur. From theosib at gmail.com Wed Jan 7 09:31:38 2009 From: theosib at gmail.com (Timothy Normand Miller) Date: Wed Jan 7 05:31:37 2009 Subject: [Open-graphics] any more news? In-Reply-To: References: Message-ID: <9871ee5f0901070631t5c39f7a2vc9c5f8e412d067ee@mail.gmail.com> Before December, we'd already had OGD1 working as a dumb framebuffer. One of the milestones reached since then is that we have HQ fully working in simulation. We have code working that does nothing but shuttle memory and engine accesses over the bridge (like what would happen if HQ were disabled). This is an important component of the VGA emulation. Petter, Howard, and I have been working on that. Meanwhile, Mark Marshal has also been making progress on BIOS and the VGA microcode that translates text to graphics. On Wed, Jan 7, 2009 at 1:14 AM, Arthur Marsh wrote: > Hi, is there any news on the development/implementation front? > > Arthur. > > _______________________________________________ > Open-graphics mailing list > Open-graphics@duskglow.com > http://lists.duskglow.com/mailman/listinfo/open-graphics > List service provided by Duskglow Consulting, LLC (www.duskglow.com) > -- Timothy Normand Miller http://www.cse.ohio-state.edu/~millerti Open Graphics Project From theosib at gmail.com Fri Jan 9 12:16:17 2009 From: theosib at gmail.com (Timothy Normand Miller) Date: Fri Jan 9 08:16:23 2009 Subject: [Open-graphics] Announcement: OGD1 HQ microcontroller fully working in real hardware Message-ID: <9871ee5f0901090916r2aed1c7ei2e083e738d2bfe2f@mail.gmail.com> We've had OGD1 working as an unaccelerated non-VGA framebuffer device working for some time now. On and off over the last month, we've been working out the major show-stopping bugs in HQ (the MIPS-like microcontroller we developed to emulate VGA). One of the major development problems was that, unlike the Xilinx tools, the Lattice synthesizer seemed to be unable to infer and properly deal with the relationship between the primary clock that runs most of HQ and the double-rate clock that ran its 3-port register file. We worked around that by duplicating the register file (possibly not actually increasing the amount of logic dedicated to it), running it on the primary clock, and extending the results forwarding mechanism to handle simultaneous read and write-back to the same register. It turns out that with the 2x clock, about the fastest we would have been able to run HQ was about 50MHz. With these changes, we can run it easily at 80MHz. Our target is at least 100MHz, but at the moment, this is acceptable for development purposes. We'll defer further optimization to a later time so that we can concentrate on the details of making VGA work. The first step is do develop just enough HQ microcode and x86 BIOS code that we can trick a PC into thinking it's a real VGA card; the objective is to be able to boot into Linux without a functional display so that we can incrementally develop the code that will make VGA actually put up a working display. -- Timothy Normand Miller http://www.cse.ohio-state.edu/~millerti Open Graphics Project From theosib at gmail.com Fri Jan 16 12:41:08 2009 From: theosib at gmail.com (Timothy Normand Miller) Date: Fri Jan 16 08:41:27 2009 Subject: [Open-graphics] Place to put OGP binaries? Message-ID: <9871ee5f0901160941k3795202bvbfdc7fa13cb9bf6f@mail.gmail.com> SVN is not a good place to store binaries (FPGA bitfiles, BIOS images, HQ microcode, etc.), but we really need a place to put some. We've been dropping them on the traversaltech.com site, which is fine, but we haven't been organized about it. Does anyone have suggestions? -- Timothy Normand Miller http://www.cse.ohio-state.edu/~millerti Open Graphics Project From petekarl at student.chalmers.se Fri Jan 16 13:46:22 2009 From: petekarl at student.chalmers.se (pk) Date: Fri Jan 16 09:46:41 2009 Subject: [Open-graphics] Place to put OGP binaries? In-Reply-To: <9871ee5f0901160941k3795202bvbfdc7fa13cb9bf6f@mail.gmail.com> References: <9871ee5f0901160941k3795202bvbfdc7fa13cb9bf6f@mail.gmail.com> Message-ID: <4970D5FE.2090800@student.chalmers.se> Timothy Normand Miller wrote: > SVN is not a good place to store binaries (FPGA bitfiles, BIOS images, > HQ microcode, etc.), but we really need a place to put some. We've > been dropping them on the traversaltech.com site, which is fine, but > we haven't been organized about it. Does anyone have suggestions? Hi, According to this: http://en.wikipedia.org/wiki/Subversion_(software) ... Subversion does handle binaries. Maybe it's a configuration issue? And RCS is also supposed to handle binaries: http://en.wikipedia.org/wiki/Revision_Control_System Best regards Peter K From theosib at gmail.com Fri Jan 16 15:48:23 2009 From: theosib at gmail.com (Timothy Normand Miller) Date: Fri Jan 16 11:48:41 2009 Subject: [Open-graphics] Place to put OGP binaries? In-Reply-To: <4970D5FE.2090800@student.chalmers.se> References: <9871ee5f0901160941k3795202bvbfdc7fa13cb9bf6f@mail.gmail.com> <4970D5FE.2090800@student.chalmers.se> Message-ID: <9871ee5f0901161248g9161bb5s36496c1bd42015b2@mail.gmail.com> You can check binaries into SVN. But when you check in updates, it doesn't do diffs, so each submission takes up a lot of space, and we really only need the latest binary. For older versions, we can rebuild from source. On Fri, Jan 16, 2009 at 1:46 PM, pk wrote: > Timothy Normand Miller wrote: >> SVN is not a good place to store binaries (FPGA bitfiles, BIOS images, >> HQ microcode, etc.), but we really need a place to put some. We've >> been dropping them on the traversaltech.com site, which is fine, but >> we haven't been organized about it. Does anyone have suggestions? > > Hi, > > According to this: > http://en.wikipedia.org/wiki/Subversion_(software) > > ... Subversion does handle binaries. Maybe it's a configuration issue? > > And RCS is also supposed to handle binaries: > http://en.wikipedia.org/wiki/Revision_Control_System > > Best regards > > Peter K > -- Timothy Normand Miller http://www.cse.ohio-state.edu/~millerti Open Graphics Project From daniel at rozsnyo.com Sat Jan 17 06:00:13 2009 From: daniel at rozsnyo.com (=?UTF-8?B?IkluZy4gRGFuaWVsIFJvenNuecOzIg==?=) Date: Sat Jan 17 02:00:39 2009 Subject: [Open-graphics] Place to put OGP binaries? In-Reply-To: <9871ee5f0901161248g9161bb5s36496c1bd42015b2@mail.gmail.com> References: <9871ee5f0901160941k3795202bvbfdc7fa13cb9bf6f@mail.gmail.com> <4970D5FE.2090800@student.chalmers.se> <9871ee5f0901161248g9161bb5s36496c1bd42015b2@mail.gmail.com> Message-ID: <4971BA3D.40502@rozsnyo.com> Could the SVN client filter out binaries at new version check-in ? Then we could replace that check-in command with a script which uploads the binaries to the download site (part of the web) and optionally generates a textfile with url to the file. That url-list will be then checked in (now with diff) to SVN. The web-application then can keep archive as wanted (e.g. 5 versions back from each person). It can visually show newest images first (with information about age of that image, the person who made it, stability information (devel/testing/stable) and optionally list of svn patches applied since previous post). The standardized location of binaries will enable an auto-update feature in future, where the customer can select its preferred stability level and the preferred developer (which kind-a correspond with the branch where the image is oriented at - e.g. one likes to hack 3D, other one video decoding and so on). Anyway, if auto-update is not wanted, it will simplify package building. Daniel Timothy Normand Miller wrote: > You can check binaries into SVN. But when you check in updates, it > doesn't do diffs, so each submission takes up a lot of space, and we > really only need the latest binary. For older versions, we can > rebuild from source. > > On Fri, Jan 16, 2009 at 1:46 PM, pk wrote: >> Timothy Normand Miller wrote: >>> SVN is not a good place to store binaries (FPGA bitfiles, BIOS images, >>> HQ microcode, etc.), but we really need a place to put some. We've >>> been dropping them on the traversaltech.com site, which is fine, but >>> we haven't been organized about it. Does anyone have suggestions? >> Hi, >> >> According to this: >> http://en.wikipedia.org/wiki/Subversion_(software) >> >> ... Subversion does handle binaries. Maybe it's a configuration issue? >> >> And RCS is also supposed to handle binaries: >> http://en.wikipedia.org/wiki/Revision_Control_System >> >> Best regards >> >> Peter K >> > > > From theosib at gmail.com Mon Jan 19 15:36:42 2009 From: theosib at gmail.com (Timothy Normand Miller) Date: Mon Jan 19 11:44:25 2009 Subject: [Open-graphics] Need help programming the XP10 with FOSS software Message-ID: <9871ee5f0901191236y33aebfa5i96679ec10d7d3b77@mail.gmail.com> Hi, everyone. We're looking for a way to program our Lattice XP10 FPGA without using Windows or commercial software. It looks like UrJTAG is a very complete system, but we can't figure out how to make it program the XP10. Does anyone know where we might start? Thanks! -- Timothy Normand Miller http://www.cse.ohio-state.edu/~millerti Open Graphics Project From cestrauss at gmail.com Mon Jan 19 16:51:43 2009 From: cestrauss at gmail.com (Cesar Strauss) Date: Mon Jan 19 13:00:27 2009 Subject: [Open-graphics] Re: Need help programming the XP10 with FOSS software In-Reply-To: <9871ee5f0901191236y33aebfa5i96679ec10d7d3b77@mail.gmail.com> References: <9871ee5f0901191236y33aebfa5i96679ec10d7d3b77@mail.gmail.com> Message-ID: Timothy Normand Miller wrote: > Hi, everyone. > > We're looking for a way to program our Lattice XP10 FPGA without using > Windows or commercial software. It looks like UrJTAG is a very > complete system, but we can't figure out how to make it program the > XP10. Does anyone know where we might start? > > Thanks! > Hello Timothy, I don't use Lattice FPGAs, but I successfully used UrJTAG to program a Virtex using SVF files and a Xilinx parallel port JTAG cable. After finding a BSDL description file for it, I used "bsdl path .", then "svf program.svf" Unfortunately, UrJTAG failed to program a Spartan 3 in another board, perhaps related to it being second in the JTAG chain. I ended using "xilprg" instead, which can even directly read Xilinx bit files. Does the Lattice tools generate SVF files? Is your JTAG cable supported by UrJTAG? Hope this helps, Cesar From urkedal at nbi.dk Mon Jan 19 17:18:08 2009 From: urkedal at nbi.dk (Petter Urkedal) Date: Mon Jan 19 13:18:44 2009 Subject: [Open-graphics] Re: Need help programming the XP10 with FOSS software In-Reply-To: References: <9871ee5f0901191236y33aebfa5i96679ec10d7d3b77@mail.gmail.com> Message-ID: <20090119221807.GA29048@eideticdew.org> On 2009-01-19, Cesar Strauss wrote: > Timothy Normand Miller wrote: > > Hi, everyone. > > > > We're looking for a way to program our Lattice XP10 FPGA without using > > Windows or commercial software. It looks like UrJTAG is a very > > complete system, but we can't figure out how to make it program the > > XP10. Does anyone know where we might start? > > > > Thanks! > > > > Hello Timothy, > > I don't use Lattice FPGAs, but I successfully used UrJTAG to program a > Virtex using SVF files and a Xilinx parallel port JTAG cable. After > finding a BSDL description file for it, I used "bsdl path .", then "svf > program.svf" > > Unfortunately, UrJTAG failed to program a Spartan 3 in another board, > perhaps related to it being second in the JTAG chain. I ended using > "xilprg" instead, which can even directly read Xilinx bit files. > > Does the Lattice tools generate SVF files? Is your JTAG cable supported > by UrJTAG? There was a little discussion off-list, and I am the one who most need this at the moment. My cable is an "FCPU-X Platform USB JTAG cable" which is supposed to be compatible with the "Xilinx Platform Cable USB". This is how far I get using UrJTAG from Subversion of today: $ sudo lsusb Bus 001 Device 004: ID 03fd:0008 Xilinx, Inc. ... $ jtag jtag> cable xpc_int firmware version = 0x044C (1100) cable CPLD version = 0x0012 (18) jtag> detect IR length: 8 Chain length: 1 Device Id: 0 (0x0000000000000000) chain.c(149) Part 0 without active instruction chain.c(200) Part 0 without active instruction chain.c(149) Part 0 without active instruction jtag> idcode Reading 0 bytes if idcode Read 01000011 11000000 00100101 00010001 00000000 00000000 00000000 00000000 The device ID looks strange. I'm not sure if it's a problem with the USB driver, the connection to the board, or lacking support for the devices on OGD1. Even if my cable can't be used, support for the XP10 to the point where we can replay an SVF would be very useful. That is, presuming the Lattice tools can generate SVF. From theosib at gmail.com Tue Jan 20 14:58:38 2009 From: theosib at gmail.com (Timothy Normand Miller) Date: Tue Jan 20 10:59:05 2009 Subject: [Open-graphics] AHA: Out-dated schematic Message-ID: <9871ee5f0901201158j4ed78626if0b3df1b5c05d695@mail.gmail.com> I kept wondering why people kept saying that the schematic online is out of date. The one on traversaltech.com's download page is the right one. The one linked to by the OGP wiki is an old one. -- Timothy Normand Miller http://www.cse.ohio-state.edu/~millerti Open Graphics Project From howard.parkin at gmail.com Tue Jan 20 15:00:16 2009 From: howard.parkin at gmail.com (howard parkin) Date: Tue Jan 20 11:00:43 2009 Subject: [Open-graphics] Re: Need help programming the XP10 with FOSS software In-Reply-To: <20090119221807.GA29048@eideticdew.org> References: <9871ee5f0901191236y33aebfa5i96679ec10d7d3b77@mail.gmail.com> <20090119221807.GA29048@eideticdew.org> Message-ID: > > Even if my cable can't be used, support for the XP10 to the point where > we can replay an SVF would be very useful. That is, presuming the > Lattice tools can generate SVF. > _______________________________________________ You can generate SVF files from JED/BIT files using the Lattice Universal File writer (ispUFW which is part of the ispVM tool suite). From josephhenryblack at gmail.com Sat Jan 24 20:15:37 2009 From: josephhenryblack at gmail.com (Josephblack) Date: Sat Jan 24 16:16:13 2009 Subject: [Open-graphics] Announcement: OGD1 HQ microcontroller fully working in real hardware In-Reply-To: <9871ee5f0901090916r2aed1c7ei2e083e738d2bfe2f@mail.gmail.com> References: <9871ee5f0901090916r2aed1c7ei2e083e738d2bfe2f@mail.gmail.com> Message-ID: I have added a one sentence announcement to the front page. http://wiki.opengraphics.org/tiki-index.php?page=HomePage improvements/suggestions welcome JHB On 10/01/2009, Timothy Normand Miller wrote: > We've had OGD1 working as an unaccelerated non-VGA framebuffer device > working for some time now. On and off over the last month, we've been > working out the major show-stopping bugs in HQ (the MIPS-like > microcontroller we developed to emulate VGA). One of the major > development problems was that, unlike the Xilinx tools, the Lattice > synthesizer seemed to be unable to infer and properly deal with the > relationship between the primary clock that runs most of HQ and the > double-rate clock that ran its 3-port register file. We worked around > that by duplicating the register file (possibly not actually > increasing the amount of logic dedicated to it), running it on the > primary clock, and extending the results forwarding mechanism to > handle simultaneous read and write-back to the same register. It > turns out that with the 2x clock, about the fastest we would have been > able to run HQ was about 50MHz. With these changes, we can run it > easily at 80MHz. Our target is at least 100MHz, but at the moment, > this is acceptable for development purposes. We'll defer further > optimization to a later time so that we can concentrate on the details > of making VGA work. The first step is do develop just enough HQ > microcode and x86 BIOS code that we can trick a PC into thinking it's > a real VGA card; the objective is to be able to boot into Linux > without a functional display so that we can incrementally develop the > code that will make VGA actually put up a working display. > > -- > Timothy Normand Miller > http://www.cse.ohio-state.edu/~millerti > Open Graphics Project > _______________________________________________ > Open-graphics mailing list > Open-graphics@duskglow.com > http://lists.duskglow.com/mailman/listinfo/open-graphics > List service provided by Duskglow Consulting, LLC (www.duskglow.com) > From josephhenryblack at gmail.com Sat Jan 24 20:47:13 2009 From: josephhenryblack at gmail.com (Josephblack) Date: Sat Jan 24 16:47:47 2009 Subject: [Open-graphics] AHA: Out-dated schematic In-Reply-To: <9871ee5f0901201158j4ed78626if0b3df1b5c05d695@mail.gmail.com> References: <9871ee5f0901201158j4ed78626if0b3df1b5c05d695@mail.gmail.com> Message-ID: Thanks for the correction. The link should be fixed now. JHB From theosib at gmail.com Sun Jan 25 11:52:35 2009 From: theosib at gmail.com (Timothy Normand Miller) Date: Sun Jan 25 07:53:10 2009 Subject: [Open-graphics] Re: Why get an original watch? In-Reply-To: References: Message-ID: <9871ee5f0901250852o3c7b021jaa6973f7993eea2f@mail.gmail.com> Did someone actually subscribe to the list just to spam us? On Sun, Jan 25, 2009 at 5:20 AM, Adele Ash wrote: > Dear Glenna, > > Looking for a Patek Phillipe watch that no one can tell from the original? You're in luck, because we have the best copies > http://search.yahoo.com/search?y=Search&p=mainlikes%2ecom&fr=sfp&ei=UTF-8 (please click on the link after "Go directly to ") > > Get two deeply discounted watches and take an extra 15% discount. > http://search.yahoo.com/search?y=Search&p=mainlikes%2ecom&fr=sfp&ei=UTF-8 (please click on the link after "Go directly to ") > > Our Patek Phillipe watches have Weights/feels and looks exactly same as original. > > Sincerely, > Mr Bray > > -- Timothy Normand Miller http://www.cse.ohio-state.edu/~millerti Open Graphics Project