Re: smp, newconfig, hdama
On Sun, 10 Aug 2003, Bari Ari wrote: ron minnich wrote: The two unknowable issues with IBM: - enable SPD - enable FLASH write they refuse to tell us. Was that with the T23? The T23 has a 830MP chipset and S3 graphics. What did they use for the keyboard controller/super I/O? that is with all IBMs. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
ron minnich wrote: On Sun, 10 Aug 2003, Bari Ari wrote: ron minnich wrote: The two unknowable issues with IBM: - enable SPD - enable FLASH write they refuse to tell us. Was that with the T23? The T23 has a 830MP chipset and S3 graphics. What did they use for the keyboard controller/super I/O? that is with all IBMs. The few IBM's I've looked at mux all the SMBus and I2C lines that are used for SPD, smart battery, keyboard controller and DDC for CRT, etc, etc. The Flash ROM's and FWH write enables tend to be controlled through GPIO's on the Super I/O or keyboard/system management controller. Bari ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
On Mon, 11 Aug 2003, Bari Ari wrote: The few IBM's I've looked at mux all the SMBus and I2C lines that are used for SPD, smart battery, keyboard controller and DDC for CRT, etc, etc. The Flash ROM's and FWH write enables tend to be controlled through GPIO's on the Super I/O or keyboard/system management controller. yes, and we can't find those enables. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
ron minnich wrote: The two unknowable issues with IBM: - enable SPD - enable FLASH write they refuse to tell us. Was that with the T23? The T23 has a 830MP chipset and S3 graphics. What did they use for the keyboard controller/super I/O? Bari ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
Stefan Reinauer wrote: There is not much you can do except gnireenigne esrever existing firmware Stefan That's it, gnireenigne esrever, perfectly legal in the US (Sony v Connectix) as long as the code is not encrypted for copy protection or content access protection (DMCA). Even then it is okay for developing interoperable software, but Judge Kaplan in the DeCSS case wrote that the law did not permit you to disseminate the information you gained from gnireenigne esrever (in the case of encryption). -Steve ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: smp, newconfig, hdama
On Tue, 5 Aug 2003, David Barr wrote: Ron, is the T23 running linuxbios? When I joined the list (though I later left) and was reading back-issues, I vaguely remember you as having said you couldn't get it on your lappy due to IBM's chipset detail stinginess; has any/all of that changed, or am I perhaps hallucinating? it's on the list, but IBM has never helped much. We'd rather work with something like the lindows system, as we think they might actually help. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
On Sun, 2003-08-10 at 19:07, ron minnich wrote: On Sun, 10 Aug 2003, Bari Ari wrote: The IBM A22-M uses the Intel 440BX chipset with the ATI Mobility Rage M3 for graphics. The 440BX is supported by LinuxBIOS and the VideoBIOS can be extracted from the flash in the unit. Find out what they use for a keyboard controller/superI/O and it may be workable. It may be one of the Renasas (Hitachi) IKAP family. The two unknowable issues with IBM: - enable SPD - enable FLASH write I thought the openbios/devbios code that i posted and made reference to had the corresponding addresses for enabling flash write for the i440bx chipset? (or is this another issue? ...i thought it was really an mtd issue.) they refuse to tell us. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios -- Roger http://www.eskimo.com/~roger/index.html ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
On Wed, 13 Aug 2003, Stefan Reinauer wrote: devbios generally knows how to do this for certain chipsets, including the 440. Unfortunately some mainboard vendors additionaly protect flash write with some undocumented GPIO circuit, so supporting the chipset alone is not enough. There is not much you can do except gnireenigne esrever existing firmware or trying out all possible GPIO lines (and combine them etc) The first is probably illegal in some countries, the second might destroy your hardware, esp. on laptops. I did in fact try all GPIO combinations on the T23 GPIO lines, as well as on the T22. Interestingly enough, some of them will reset the laptop. But I never found the combination, after much work, that would open up SPD or enable FLASH write. Some of these designers put an extra flipflop in such that the control bits are sticky -- you can turn the SPD off, for example, but not back on. We found out how to do this type of thing on the L440GX, and the hardware was not simple. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
* Roger [EMAIL PROTECTED] [030811 10:02]: The two unknowable issues with IBM: - enable SPD - enable FLASH write I thought the openbios/devbios code that i posted and made reference to had the corresponding addresses for enabling flash write for the i440bx chipset? (or is this another issue? ...i thought it was really an mtd issue.) devbios generally knows how to do this for certain chipsets, including the 440. Unfortunately some mainboard vendors additionaly protect flash write with some undocumented GPIO circuit, so supporting the chipset alone is not enough. There is not much you can do except gnireenigne esrever existing firmware or trying out all possible GPIO lines (and combine them etc) The first is probably illegal in some countries, the second might destroy your hardware, esp. on laptops. Stefan -- Architecture Team SuSE Linux AG ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
Why not try 64 bit? Yinghai Lu -- : ron minnich [mailto:[EMAIL PROTECTED] : 200385 16:20 : YhLu : [EMAIL PROTECTED] : Re: smp, newconfig, hdama On Tue, 5 Aug 2003, YhLu wrote: You need to change _RAMBASE to 0x4000. thank you, that did it. I now have an HDAMA running as a 2-cpu bproc slave node, in 32-bit mode, with an IBM T23 laptop as the master. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
On Tue, 5 Aug 2003, YhLu wrote: You need to change _RAMBASE to 0x4000. thank you, that did it. I now have an HDAMA running as a 2-cpu bproc slave node, in 32-bit mode, with an IBM T23 laptop as the master. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: smp, newconfig, hdama
Ron, is the T23 running linuxbios? When I joined the list (though I later left) and was reading back-issues, I vaguely remember you as having said you couldn't get it on your lappy due to IBM's chipset detail stinginess; has any/all of that changed, or am I perhaps hallucinating? I have access to a fairly largish supply of A22-M series, and as president of my local student ACM chapter (named Beta Iota Tau, nerdily enough) am desparately searching for more projects for the upcoming semester/year. Short Summary: Does linuxBios work on your T23? If so, can I have all the juicy (or at least some moderately damp) details? TIA, -David H. Barr. thank you, that did it. I now have an HDAMA running as a 2-cpu bproc slave node, in 32-bit mode, with an IBM T23 laptop as the master. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
David Barr wrote: I have access to a fairly largish supply of A22-M series, and as president of my local student ACM chapter (named Beta Iota Tau, nerdily enough) am desparately searching for more projects for the upcoming semester/year. Also take a look at the flash. If it's soldered on, you'll need to have a socket put on before you should dare trying to flash it. Use one laptop to get a working image, then you can flash the rest with their flash still soldered on. Also, put a layer of foam under the keyboard for when you lay it down in its socket in order to be able to turn the laptop on. The underside of those keyboards is metal and can short out the motherboard. I never thought to look and learned that lesson the hard way, one dead T23. Let me know if you need to find a place to have a socket put on for the flash. - Adam Agnew ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: smp, newconfig, hdama
On Tue, 5 Aug 2003, YhLu wrote: Why not try 64 bit? that's next. I am waiting for something. You'll see :-) ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
smp, newconfig, hdama
fails. ah well it was too good to last. LinuxBIOS is saying that CPU #1 did not initialize. just FYI, this is doubtless a setup problem. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
