Hello! Joe, just for fun I ran the supported boxes on your website past Amazon and surprisingly enough it found two of them so far.
The first one was the original RCA RM4100, and the imagery there confirmed what it was originally programmed to do. There are several available there at the usual inflated prices.
The second one was the RCA IP1150. I gave Amazon the full name including the Akimbo service name, according to them the box was originally conceived of being a video on demand player for that service. I've got a strange feeling that the service does not exist anymore.
This is what Amazon found http://tinyurl.com/akimbo1150 .
I am tempted to buy it, if only for the fun of it.
Now the question of the week, is the CoreBoot target for this fellow complete? And its obvious back up, what about TV out for this one, under it? According to your site, for the RM4100 it has been accomplished.
----- Gregg C Levine gregg.drwho8@gmail.com "This signature fought the Time Wars, time and again."
On 05/27/2010 03:00 AM, Gregg Levine wrote:
Hello! Joe, just for fun I ran the supported boxes on your website past Amazon and surprisingly enough it found two of them so far.
The first one was the original RCA RM4100, and the imagery there confirmed what it was originally programmed to do. There are several available there at the usual inflated prices.
The second one was the RCA IP1150. I gave Amazon the full name including the Akimbo service name, according to them the box was originally conceived of being a video on demand player for that service. I've got a strange feeling that the service does not exist anymore.
This is what Amazon found http://tinyurl.com/akimbo1150 .
I am tempted to buy it, if only for the fun of it.
Now the question of the week, is the CoreBoot target for this fellow complete? And its obvious back up, what about TV out for this one, under it? According to your site, for the RM4100 it has been accomplished.
Hello Gregg, The IP1150 has a i854 chipset. Everything else on the board is supported by coreboot but the i854. This is my next coreboot port (coming soon). There is a good site based on the IP1101, but they have the same board (Thomson codename mambux). You can check out the site for more info: http://www.mambux.de/
On Thu, May 27, 2010 at 4:41 PM, Joseph Smith joe@settoplinux.org wrote:
On 05/27/2010 03:00 AM, Gregg Levine wrote:
Hello! Joe, just for fun I ran the supported boxes on your website past Amazon and surprisingly enough it found two of them so far.
The first one was the original RCA RM4100, and the imagery there confirmed what it was originally programmed to do. There are several available there at the usual inflated prices.
The second one was the RCA IP1150. I gave Amazon the full name including the Akimbo service name, according to them the box was originally conceived of being a video on demand player for that service. I've got a strange feeling that the service does not exist anymore.
This is what Amazon found http://tinyurl.com/akimbo1150 .
I am tempted to buy it, if only for the fun of it.
Now the question of the week, is the CoreBoot target for this fellow complete? And its obvious back up, what about TV out for this one, under it? According to your site, for the RM4100 it has been accomplished.
Hello Gregg, The IP1150 has a i854 chipset. Everything else on the board is supported by coreboot but the i854. This is my next coreboot port (coming soon). There is a good site based on the IP1101, but they have the same board (Thomson codename mambux). You can check out the site for more info: http://www.mambux.de/
-- Thanks, Joseph Smith Set-Top-Linux www.settoplinux.org
Hello! Joe, just for fun I pulled over my current Coreboot collection over to my Slackware 11.0 system. After some interesting finagling, namely reconstructing the tree, and some interesting Subversion issues that aren't relevant here, it arrived.
I also checked to see if I could build your code for producing TV out on the first target. On Slackware 11.0 it built without comments (no warnings issued) but on Slackware 12.1 which is what my primary system is running there were a remarkable load of errors issued, which resulted in it not being built.
Strangely enough it happened towards the end of the cycle of steps to build it. I ran the script program to create a text file containing those events after trying it again. If you want I can send you the script output directly as I do not believe its relevant to the Coreboot list.
Now the important question, what distribution do you use for your work? It might explain why I received those errors.
Incidentally about the box that's the subject of this thread, I am still making my way through my decision. I'll know more by the end of the coming week. ----- Gregg C Levine gregg.drwho8@gmail.com "This signature fought the Time Wars, time and again."
I should mention that in parallel universe I work in there's lots of weird things going on with gcc. It's a nightmare, highly threaded program that blows up with gcc 4.4 ... so we're not alone.
ron
On Sat, May 29, 2010 at 2:37 PM, ron minnich rminnich@gmail.com wrote:
I should mention that in parallel universe I work in there's lots of weird things going on with gcc. It's a nightmare, highly threaded program that blows up with gcc 4.4 ... so we're not alone.
ron
Hello! I see your point, Ron. I do indeed. However I did some final analysis on the output that I can either post here, on the Coreboot list if you're curious, or not and directly to Joe, the explosion of complaints happens on my Slackware 12.1 system, simply because that system is running a later version of the PCI libraries. And the program builds properly on the other one running Slackware 11.0 because the PCI libraries are older. I am referring to libpci.a here. According to the lists of installed things, its 2.2.10 for the 12.1 system.
To be honest I really don't know why this is happening: root@jimkirk2:/usr/src/lobos/stb/trunk/ip1000_tvout# make gcc -O2 -g -Wall -c -o main.o main.c gcc -O2 -g -Wall -c -o gmch.o gmch.c gcc -O2 -g -Wall -c -o i2c.o i2c.c gcc -O2 -g -Wall -c -o cx2587x.o cx2587x.c gcc -O2 -g -Wall -c -o i830_video.o i830_video.c gcc -O2 -g -Wall -c -o video_modes.o video_modes.c gcc -o ip1000-tvout main.o gmch.o i2c.o cx2587x.o i830_video.o video_modes.o -lp ci /usr/lib/gcc/i486-slackware-linux/4.2.3/../../../libpci.a(names.o): In function `pci_load_name_list': names.c:(.text+0x46b): undefined reference to `gzopen' names.c:(.text+0x4f3): undefined reference to `gzgets' names.c:(.text+0x54a): undefined reference to `gzeof' names.c:(.text+0x612): undefined reference to `gzclose' names.c:(.text+0x783): undefined reference to `gzclose' names.c:(.text+0x975): undefined reference to `gzerror' names.c:(.text+0x992): undefined reference to `gzclose' names.c:(.text+0xa27): undefined reference to `gzopen' names.c:(.text+0xaa0): undefined reference to `zError' collect2: ld returned 1 exit status make: *** [ip1000-tvout] Error 1
And not on the Slackware 11.0 system. It is running the 2.2.3 release. This is why Joe I asked earlier whose distribution you chose and what not. For all I know the code in question may have been written not by you, but by that fellow who discovered how to make TV out work on that board. There's an excellent write up on www.settoplinux.org that goes into details leading up to the code, and then the location where it lives.
Depending on when I get the hardware here, it might be one, the one I asked you about earlier even, or the regular participant, the IP1000 one, we'll be able to resolve this sooner.
----- Gregg C Levine gregg.drwho8@gmail.com "This signature fought the Time Wars, time and again."
On 05/29/2010 02:54 PM, Gregg Levine wrote:
On Sat, May 29, 2010 at 2:37 PM, ron minnichrminnich@gmail.com wrote:
I should mention that in parallel universe I work in there's lots of weird things going on with gcc. It's a nightmare, highly threaded program that blows up with gcc 4.4 ... so we're not alone.
ron
Hello! I see your point, Ron. I do indeed. However I did some final analysis on the output that I can either post here, on the Coreboot list if you're curious, or not and directly to Joe, the explosion of complaints happens on my Slackware 12.1 system, simply because that system is running a later version of the PCI libraries. And the program builds properly on the other one running Slackware 11.0 because the PCI libraries are older. I am referring to libpci.a here. According to the lists of installed things, its 2.2.10 for the 12.1 system.
To be honest I really don't know why this is happening: root@jimkirk2:/usr/src/lobos/stb/trunk/ip1000_tvout# make gcc -O2 -g -Wall -c -o main.o main.c gcc -O2 -g -Wall -c -o gmch.o gmch.c gcc -O2 -g -Wall -c -o i2c.o i2c.c gcc -O2 -g -Wall -c -o cx2587x.o cx2587x.c gcc -O2 -g -Wall -c -o i830_video.o i830_video.c gcc -O2 -g -Wall -c -o video_modes.o video_modes.c gcc -o ip1000-tvout main.o gmch.o i2c.o cx2587x.o i830_video.o video_modes.o -lp ci /usr/lib/gcc/i486-slackware-linux/4.2.3/../../../libpci.a(names.o): In function `pci_load_name_list': names.c:(.text+0x46b): undefined reference to `gzopen' names.c:(.text+0x4f3): undefined reference to `gzgets' names.c:(.text+0x54a): undefined reference to `gzeof' names.c:(.text+0x612): undefined reference to `gzclose' names.c:(.text+0x783): undefined reference to `gzclose' names.c:(.text+0x975): undefined reference to `gzerror' names.c:(.text+0x992): undefined reference to `gzclose' names.c:(.text+0xa27): undefined reference to `gzopen' names.c:(.text+0xaa0): undefined reference to `zError' collect2: ld returned 1 exit status make: *** [ip1000-tvout] Error 1
And not on the Slackware 11.0 system. It is running the 2.2.3 release. This is why Joe I asked earlier whose distribution you chose and what not. For all I know the code in question may have been written not by you, but by that fellow who discovered how to make TV out work on that board. There's an excellent write up on www.settoplinux.org that goes into details leading up to the code, and then the location where it lives.
Depending on when I get the hardware here, it might be one, the one I asked you about earlier even, or the regular participant, the IP1000 one, we'll be able to resolve this sooner.
Hello Gregg, We used Fedora and Debian to develop the software. Maybe it needs an update, I haven't tried to build it in a while. There was a collaboration of people who figured it out (see README) but one person that actually wrote the code. By the time you get your box I may already have tv-out working at boot (in coreboot 99% there) so this program would only be needed to make adjustments if desired. Hope that helps.
Hi,
On 29.05.2010 20:54, Gregg Levine wrote:
system is running a later version of the PCI libraries. And the program builds properly on the other one running Slackware 11.0 because the PCI libraries are older. I am referring to libpci.a here. According to the lists of installed things, its 2.2.10 for the 12.1 system.
To be honest I really don't know why this is happening: root@jimkirk2:/usr/src/lobos/stb/trunk/ip1000_tvout# make gcc -O2 -g -Wall -c -o main.o main.c gcc -O2 -g -Wall -c -o gmch.o gmch.c gcc -O2 -g -Wall -c -o i2c.o i2c.c gcc -O2 -g -Wall -c -o cx2587x.o cx2587x.c gcc -O2 -g -Wall -c -o i830_video.o i830_video.c gcc -O2 -g -Wall -c -o video_modes.o video_modes.c gcc -o ip1000-tvout main.o gmch.o i2c.o cx2587x.o i830_video.o video_modes.o -lp ci /usr/lib/gcc/i486-slackware-linux/4.2.3/../../../libpci.a(names.o): In function `pci_load_name_list': names.c:(.text+0x46b): undefined reference to `gzopen' names.c:(.text+0x4f3): undefined reference to `gzgets' names.c:(.text+0x54a): undefined reference to `gzeof' names.c:(.text+0x612): undefined reference to `gzclose' names.c:(.text+0x783): undefined reference to `gzclose' names.c:(.text+0x975): undefined reference to `gzerror' names.c:(.text+0x992): undefined reference to `gzclose' names.c:(.text+0xa27): undefined reference to `gzopen' names.c:(.text+0xaa0): undefined reference to `zError' collect2: ld returned 1 exit status make: *** [ip1000-tvout] Error 1
Haha, the infamous libpci dependency on libz strikes again. Either recompile libpci without gzip compression support, or add -lz to the linker options in the makefile. That's it.
An alternative would be to use the huge chunk of code in the flashrom makefile which deals with this ugly wart of libpci. There's no need to write such makefile code from scratch because flashrom already handles it, and it works on pretty much every platform out there (DOS/Linux/*BSD/Solaris/MacOSX).
Regards, Carl-Daniel
On Sat, May 29, 2010 at 8:00 PM, Carl-Daniel Hailfinger c-d.hailfinger.devel.2006@gmx.net wrote:
Hi,
On 29.05.2010 20:54, Gregg Levine wrote:
system is running a later version of the PCI libraries. And the program builds properly on the other one running Slackware 11.0 because the PCI libraries are older. I am referring to libpci.a here. According to the lists of installed things, its 2.2.10 for the 12.1 system.
To be honest I really don't know why this is happening: root@jimkirk2:/usr/src/lobos/stb/trunk/ip1000_tvout# make gcc -O2 -g -Wall -c -o main.o main.c gcc -O2 -g -Wall -c -o gmch.o gmch.c gcc -O2 -g -Wall -c -o i2c.o i2c.c gcc -O2 -g -Wall -c -o cx2587x.o cx2587x.c gcc -O2 -g -Wall -c -o i830_video.o i830_video.c gcc -O2 -g -Wall -c -o video_modes.o video_modes.c gcc -o ip1000-tvout main.o gmch.o i2c.o cx2587x.o i830_video.o video_modes.o -lp ci /usr/lib/gcc/i486-slackware-linux/4.2.3/../../../libpci.a(names.o): In function `pci_load_name_list': names.c:(.text+0x46b): undefined reference to `gzopen' names.c:(.text+0x4f3): undefined reference to `gzgets' names.c:(.text+0x54a): undefined reference to `gzeof' names.c:(.text+0x612): undefined reference to `gzclose' names.c:(.text+0x783): undefined reference to `gzclose' names.c:(.text+0x975): undefined reference to `gzerror' names.c:(.text+0x992): undefined reference to `gzclose' names.c:(.text+0xa27): undefined reference to `gzopen' names.c:(.text+0xaa0): undefined reference to `zError' collect2: ld returned 1 exit status make: *** [ip1000-tvout] Error 1
Haha, the infamous libpci dependency on libz strikes again. Either recompile libpci without gzip compression support, or add -lz to the linker options in the makefile. That's it.
An alternative would be to use the huge chunk of code in the flashrom makefile which deals with this ugly wart of libpci. There's no need to write such makefile code from scratch because flashrom already handles it, and it works on pretty much every platform out there (DOS/Linux/*BSD/Solaris/MacOSX).
Regards, Carl-Daniel
Hello! It does make sense. I'll examine the makefile for the for the primary system and pass the changes along to the secondary while I am at it. Thank you Carl-Daniel.
As for why I am interested in these systems, it turns out they are perfect for displaying digital data directly. I'll need to track down the data on the NB and SB portions of the boards used in the systems, but I'll cross that bridge when I need to.
Incidentally Joe, it turns out that Intel provides a fairly complete Linux developers kit for the target inside the Akimbo platform. It's referenced on the pages belonging to the fellow who helped you with the TV out issue for the first platform. Such as the page discussing adding a serial port to the Akimbo platform. ----- Gregg C Levine gregg.drwho8@gmail.com "This signature fought the Time Wars, time and again."
-
Carl-Daniel Hailfinger -
Gregg Levine -
Joseph Smith -
ron minnich