The "stable" ftape driver is incompatible with 2.2.x kernels. You need
the latest "unstable" version instead.
On Sun, Nov 14, 1999 at 08:52:37AM -0500, Robert L. Becker Jr. wrote:
> I am a new list subscriber hoping for help to solve problems(s) installing
> ftape. Anyone not interested in background, please save some time and skip
> on to the "problem" paragraphs.
>
> Background -- I have Linux 2.2.13-4mdk (obtained as the current commercial
> release of MacMillan/Mandrake/RedHat titled Linux 6.5) running on an HP
> Pentium II, equipped with an Iomega Ditto Max (*might* be the
> 'Professional' model) tape drive using a separate ISA floppy controller
> from Iomega. (The controller is not the one that shipped with the drive a
> year ago -- Iomega had to send a separate one that was suitable for use
> with Windows NT that also runs on my host.) In experimenting with kernels
> to support various devices, I eventually compiled ftape from the Mandrake
> distribution into a kernel, noting the controller parameters when prompted
> for them under the CONFIG_FT_ALT_FDC submenus. This is ftape v3.04d
> 25/11/97 for Linux 2.2.13-4mdk. For good measure, I put the same
> parameters in a lilo.conf 'append' statement. The kernel booted ok. The
> *ft* devices showed up ok in /dev. Trying to access *ft0 via commands like
> 'mt -f /dev/rft0 status' always failed on the first attempt after booting
> with a message something like "device not configured", but succeeded on
> second and later attempts. Trying to actually work with the tape drive
> (e.g. mt -f /dev/rft0 rewind, or tar -f /dev/rft0 /etc) always failed. I
> also found messages like the following showing up in /var/log/messages:
>
> Nov 11 16:45:02 rlb kernel: ============ unknown vendor id ===========
> Nov 11 16:45:02 rlb kernel: A new, yet unsupported tape drive is found
> Nov 11 16:45:02 rlb kernel: Please report the following values:
> Nov 11 16:45:02 rlb kernel: Vendor id : 0x8885
> Nov 11 16:45:02 rlb kernel: Wakeup method : Colorado
> Nov 11 16:45:02 rlb kernel: And a description of your tape drive
> Nov 11 16:45:02 rlb kernel: to the ftape maintainer
> Nov 11 16:45:02 rlb kernel: ==========================================.
>
> So I headed off to the ftape home page and found the apparent solution,
> i.e. that the Ditto Max and Ditto Max Pro are only supported in ftape 4.x
> releases. I downloaded ftape 4.02, the latest stable version of the
> drivers. Now I'm at last to the point for describing my main problem...
>
> Problem #1 -- After unpacking the distribution and editing MCONFIG by
> commenting out these lines
>
> #FDC_OPT += -DCONFIG_FT_STD_FDC_$(STD_DEV) \
> # -DCONFIG_FT_INT_$(STD_DEV)=1
>
> and uncommenting or adding these lines
>
> FDC_OPT = -DCONFIG_FT_FDC_BASE_$(82078_DEV)=0x370 \
> -DCONFIG_FT_FDC_IRQ_$(82078_DEV)=6 \
> -DCONFIG_FT_FDC_DMA_$(82078_DEV)=2 \
> -DCONFIG_FT_NR_BUFFERS=3
> -DCONFIG_FT_INT_$(82078_DEV)=1
> FDC_BASE = 0x360
> FDC_IRQ = 7
> FDC_DMA = 3
>
> I run make only to have it fail with
>
> fdc-io.c: In function `fdc_interrupt_wait':
> fdc-io.c:433: structure has no member named `timeout'
> fdc-io.c:454: structure has no member named `timeout'
> make[2]: *** [fdc-io.o] Error 1
> make[2]: Leaving directory
> `/usr/local/packages/ftape/ftape-4.02/ftape/lowlevel'
> make[1]: *** [modules] Error 2
> make[1]: Leaving directory `/usr/local/packages/ftape/ftape-4.02/ftape'
> make: *** [all] Error 2
>
> The offending references are in lines like:
>
> current->timeout = jiffies + (1000 * time + FT_USPT - 1) / FT_USPT;
>
> and
>
> current->timeout = 0; /* interrupt hasn't cleared this */
>
> I've tried tracking down the declarations of 'current' and it's structure,
> but no luck. In any case, this smells a lot more like a problem with
> mismatched package versions than like something I could hope to fix at the
> source code level. For info, my installed gcc (shipped with the OS) is as
> follows:
>
> [root@rlb lowlevel]# gcc --version
> pgcc-2.91.66
>
> Also for info, the 'make' failure is the same for ftape 4.02 distributions
> downloaded either from the ftap homepage or from the UNC Metalab site. Is
> this make failure a known installation hitch? Solution?
>
> Problem #2 -- I thought I might find answer to Problem #1 in the available
> documentation. I downloaded same (ftape-4.x-doc-1.03) from UNC, but can't
> get *it* to install. The configure script fails as follows:
>
> ./configure --prefix=/usr/local/packages/ftape/ftapedocinstall
>
> ...lots of info with no apparent problem, then...
>
> checking for sgml2info... no
> checking whether is broken... yes
> checking for own version ... ...
>
> and the script hangs. I have used autoconfigurator without problem for one
> other package (opie one-time password utilities) since setting up my
> system. Will be grateful for ideas on how to make this work for ftape.
>
> Thanks in advance...
>
> Robert L. Becker, Jr.
> Col, USAF, MC
> Department of Cellular Pathology
> Armed Forces Institute of Pathology
> Washington, DC 20306-6000
> 202-782-1573
>
>
--
Bob Nielsen Internet: [EMAIL PROTECTED]
Tucson, AZ AMPRnet: [EMAIL PROTECTED]
DM42nh http://www.primenet.com/~nielsen
