Promise pdc20276 patch and FYI SLE
Ralf, I am presently building a jumbo file/print server solution, based on a motherboard featuring a pdc20276 and an AMD XP1800+ As per this page 'http://www.cs.helsinki.fi/linux/linux-kernel/2002-11/0039.html' where I found your patch. I am using SuSE 7.3, I have many solutions running SuSE. I could not get the drivers patched to work with 7.3. So I waited unitl 8.0 and guess what, no support for the pdc20276, only support up to the 20275. The kernel version (SuSE 8.0/2.4.18) is very close to the version kernel in which your patch works. So I patched the drivers. I had a few problems and it nearly worked. At this point I had had a look at 8.0 well ??? I know 7.3 inside out, I have 20 installs (3 svrs and 17 linux wkstns and 50 windows machines working with the svrs) working with this version.. Next I scrapped 8.0, re-installed 7.3 used SuSE 8.0 kernel source. Copied the patched drivers to the correct locations. Recompiled the kernel. It did not work the first time as the booting kernel could not find the 'reiserfs' module, recompiled again to put reiserfs support into the kernel. At first boot I got a kernel panic message, rebooted and used 'linux ide=reverse' and everything came to life. I shutdown, then moved the hdd (maxtor udma133 80g) from the normal ide contoller to the PDC20276 and everything came to life. An 'lspci' indicates a 'promise raid controller' not correctly indentified but at at the 5275 identifying address. 'hdparm' is indicating some really excellent figures, these figures are in-line with what I would expect in a udma133 environment and nearly as good as a u160 scsi. So the udma133 is working. This server will be a fast machine, for the price. The only thing left for me to try is to get the PDC20276s hardware Raid 1 working. Then with a similar motherboard test 8.0 with the patched drivers, I did find an error in the last entry patched to 'pdc202xx.c', this will almost certainly explain why it nearly work as I discussed above. This error was not detected during compilation, but was crucial to functionality. With SuSE 8.0 I still have some questions, but I will become more familiar with the distribution before I implement a solution for a client based on the 8.0. Thanks for your patch, the good will, expertise, and technical excellence of our opensource community. Regards Paul Ketelaar Mackay, Queensland, Australia.
participants (1)
-
Paul Ketelaar