Frequently asked questions concerning UltraStor SCSI adaptors. There have been many postings of the form "SLS can't find my UltraStor SCSI adaptor". The same problem also occurs with other available installation disks. The following concerns the '14F and '34F adaptors, and may or may not apply to the '24F. The problem is that there is a driver for the Accent serial board built into the SLS and the other distribution disks I know of. During the early part of booting, Linux does a "roll call" to see what kind of hardware options are present. Typically an optional board has logic to give a predefined reply to a particular I/O command. The driver can issue this command to the various I/O addresses where its board might be, and see whether it gets the expected reply. If it does get that reply, it can set itself up to use that address in later access to the board. What happens is that the Accent driver issues some kind of command to an address in the 330h-33fh range in the hope that an Accent board might reply. This hits the UltraStor 14F/34F and hangs it. Following this, the UltraStor is not detected, and booting completes without any hard disk drive. Installation is obviously not possible. Avoiding this with the '14F is simple - just change the jumpers to use a different I/O address. This didn't seem to work with the '34F, as the address didn't change when the jumpers were added. The solution, according to UltraStor support, is to make the jumper changes on the same pins (1-2 and 3-4) of JP3 instead of JP4. This does change the address. I don't know whether this will be changed to match the manual, or vice versa. This works on this machine and for others I know of. Once you have the system installed, it is simple to rebuild the kernel without the offending driver by answering "n" at the obvious point in the "make config" process. You can then use the default 330h I/O address without further problems. The alternative is to get a kernel that works with the default I/O address from someone. If you have more questions about this, let me know. B. D. Elliott bde@halcyon.com (Seattle)
| Vladislav Bolkhovitin | Re: Integration of SCST in the mainstream Linux kernel |
| Linus Torvalds | Re: 2.6.25-git2: BUG: unable to handle kernel paging request at ffffffffffffffff |
| S.Çağlar | Rescheduling interrupts |
| Andi Kleen | Re: [patch] Add basic sanity checks to the syscall execution patch |
git: | |
| Jon Smirl | Re: Figured out how to get Mozilla into git |
| Matt McCutchen | Multiple checkouts of the same repository |
| Willy Tarreau | Multiple working trees with GIT ? |
| Linus Torvalds | Re: git + ssh + key authentication feature-request |
| Richard Stallman | Real men don't attack straw men |
| Adam Getchell | Re: About Xen: maybe a reiterative question but .. |
| Girish Venkatachalam | Thinkpad t61 OpenBSD support? |
| carlopmart | About Xen: maybe a reiterative question but .. |
| Jeff Kirsher | [PATCH 1/3] e1000e: add support for the 82567LM-4 device |
| slavon | Re: e1000_clean_tx_irq: Detected Tx Unit Hang - it's bug? |
| Hugh Dickins | Re: [bug?] tg3: Failed to load firmware "tigon/tg3_tso.bin" |
| Eliezer Tamir | Re: [PATCH][BNX2X] added register coments |
