Greetings, linux-tape folks,
[NOTE: I still don't have confirmation on my subscription to the
linux-tape list, so please cc me in replies.]
A short summary:
I wanted to compile ftape 4.03pre2 with the Linux 2.2.1 kernel. The
compilation died with this error while trying to compile `fdc-io.c'
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/src/modules/ftape-4.03pre2/ftape/lowlevel'
make[1]: *** [modules] Error 2
make[1]: Leaving directory `/usr/src/modules/ftape-4.03pre2/ftape'
make: *** [all] Error 2
The variable structure trying to read `timeout' is current->timeout
This is truly a show-stopper bug, so I decided to see what I could
discover.
THE LONGER REPORT
After my first post about the problem, Perry Gilfillan <[EMAIL PROTECTED]>
replied by telling me that a few others had been having troubles. He
pointed me as far as linux/include/asm-i386/current.h which sets up
current as `struct task_struct'
With some further digging I found that linux/include/linux/sched.h
sets up the `struct task_struct' I had a 2.0.x copy of sched.h sitting
around and found the presence of `timeout'.
Next step was to find out where it went. So I dug through the kernel
patch browser and found this entry in the patch-2.1.127 file.
=-=-=-=-=-=-=-=-=-=-=-
patch-2.1.127 linux/include/linux/sched.h
Date: Sat Nov 7 11:31:11 1998
=-=-=-=-=-=-=-=-=-=-=-
diff -u --recursive --new-file v2.1.126/linux/include/linux/sched.h
linux/include/linux/sched.h
@@ -119,9 +119,10 @@
extern void show_state(void);
extern void trap_init(void);
+#define MAX_SCHEDULE_TIMEOUT LONG_MAX
+extern signed long FASTCALL(schedule_timeout(signed long timeout));
asmlinkage void schedule(void);
-
/*
* Open file table structure
*/
@@ -258,7 +259,7 @@
struct task_struct **tarray_ptr;
struct wait_queue *wait_chldexit; /* for wait4() */
- unsigned long timeout, policy, rt_priority;
+ unsigned long policy, rt_priority;
unsigned long it_real_value, it_prof_value, it_virt_value;
unsigned long it_real_incr, it_prof_incr, it_virt_incr;
struct timer_list real_timer;
@@ -348,7 +349,7 @@
/* pidhash */ NULL, NULL, \
/* tarray */ &task[0], \
/* chld wait */ NULL, \
-/* timeout */ 0,SCHED_OTHER,0,0,0,0,0,0,0, \
+/* timeout */ SCHED_OTHER,0,0,0,0,0,0,0, \
/* timer */ { NULL, NULL, 0, 0, it_real_fn }, \
/* utime */ {0,0,0,0},0, \
/* per CPU times */ {0, }, {0, }, \
@@ -457,7 +458,10 @@
extern void FASTCALL(__wake_up(struct wait_queue ** p, unsigned int mode));
extern void FASTCALL(sleep_on(struct wait_queue ** p));
+extern void FASTCALL(sleep_on(struct wait_queue ** p));
extern void FASTCALL(interruptible_sleep_on(struct wait_queue ** p));
+extern long FASTCALL(interruptible_sleep_on_timeout(struct wait_queue ** p,
+ signed long timeout));
extern void FASTCALL(wake_up_process(struct task_struct * tsk));
#define wake_up(x) __wake_up((x),TASK_UNINTERRUPTIBLE |
TASK_INTERRUPTIBLE)
=-=-=-=-=-=-=-=-=-=-
That tracks down the change. So that would mean that anyone trying to
compile ftape with 2.1.127 and after would have hit the compilation
error.
The last thing I did was to see if there was any indication why the
file was changed. I found this in the patch summary files on
www.linuxhq.com ...
=-=-=-=-=-=-=-=-=-=-
Ok,
after two fairly hectic weeks for me, 2.1.127 is finally out there.
This kernel does:
various small but important networking fixes from Davem (thanks). One
of them is the "anti-nagle" bit to allow programs that know what they
are doing to avoid nagling by telling the kernel so. This is mainly
things like Web servers and ftp-servers that can use this option
together with "sendfile()".
scheduling timeout interface change: the new interface is much more
logical than the old one, and allows us to get the jiffies wrap-around
case right. Thanks to Andrea Arcangeli.
various driver updates: specialix, sonycd,
Memory management fixups. Handle out-of-memory conditions correctly,
and handle high memory load much more gracefully.
sparc and PowerPC architecture updates
3c509 SMP fix, tlan PCI probe update.
scsi driver updates: ncr53c8xx, aic7xxx, dc390
filesystem updates: autofs, hfs, umsdos
Go, test, be happy,
Linus
=-=-=-=-=-=-=-=-=-=-=-
It looks like the second point is the important one about the
scheduling timeout interface change. I have hit the extent of my
kernel hacking knowledge (actually I exceeded it a long time ago) I
think this narrows down where the problem resides, but I have no clue
as to how to fix it. A glance at the linux/kernel/sched.c shows there
are some routines there that might be helpful. But as I have said, I
don't know much about either ftape or the kernel.
I hope this helps.
David
mailto:[EMAIL PROTECTED]
