Howdy,
I've been playing around with the CVS tree trying to get the TigerMPX stuff to compile.
At first I got assembler warnings because AMD768_DEV was not defined. I found that AMD766_DEV was defined, so I changed it. That got a little further. It was set to 0x3800 yet my lspci -vvv shows my PCI bridge at 0x2000. Is that because the Tyan BIOS moved it?
Then I got a compile failure on the ide code. It tries to include <ide.h> which doesn't exist. I #if 0'd around this because I use a different controller anyway. That got further along.
Then, after all the 768 stuff compiled, I got link failures as it was trying to link amd766 functions instead of the 768 functions.
This all leads me to believe that the code in CVS is not really functional to begin with. Either that, or I am doing something really stupid, which is entirely possible. ;) Is there newer code that someone can send me, and/or commit to CVS?
Last but not least, Sourceforge seems to be having a lot of CVS problems. It is slow when it works, but it doesn't work much of the time I try to use it. I haven't been able to update in a few days, it just drops my connections.
Thanks & Regards,
Jeff
Greetings,
I'm not honestly sure where the problems are coming from unless you got a corrupted checkout. I will be in the office next week, and will do a fresh checkout and see if I can duplicate the problem.
G'day, sjames
On Fri, 8 Aug 2003, Jeff Noxon wrote:
Howdy,
I've been playing around with the CVS tree trying to get the TigerMPX stuff to compile.
At first I got assembler warnings because AMD768_DEV was not defined. I found that AMD766_DEV was defined, so I changed it. That got a little further. It was set to 0x3800 yet my lspci -vvv shows my PCI bridge at 0x2000. Is that because the Tyan BIOS moved it?
Then I got a compile failure on the ide code. It tries to include <ide.h> which doesn't exist. I #if 0'd around this because I use a different controller anyway. That got further along.
Then, after all the 768 stuff compiled, I got link failures as it was trying to link amd766 functions instead of the 768 functions.
This all leads me to believe that the code in CVS is not really functional to begin with. Either that, or I am doing something really stupid, which is entirely possible. ;) Is there newer code that someone can send me, and/or commit to CVS?
Last but not least, Sourceforge seems to be having a lot of CVS problems. It is slow when it works, but it doesn't work much of the time I try to use it. I haven't been able to update in a few days, it just drops my connections.
Thanks & Regards,
Jeff _______________________________________________ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
-
Jeff Noxon -
steven james