Kernel-2.4.7.pre5

Kernel-2.4.7.pre5

[Kaoru Fukui]

Hi!

I had this problem in linuxppc_2_4 bk tree's linux-2.4.7.pre5.
Is my fix correct ?

Kaoru

------------- fix ---------------
diff -urN base/include/asm-ppc/hardirq.h fixed/include/asm-ppc/hardirq.h
--- base/include/asm-ppc/hardirq.h	Tue Jul 10 14:42:03 2001
+++ fixed/include/asm-ppc/hardirq.h	Tue Jul 10 20:26:26 2001
@@ -20,6 +20,7 @@
 	unsigned int __local_bh_count;
 	unsigned int __syscall_count;
 	unsigned int __last_jiffy_stamp;
+	struct task_struct * __ksoftirqd_task;
 } ____cacheline_aligned irq_cpustat_t;

 #include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above
*/

--------------------------------
-------- problem -------------
gcc -D__KERNEL__ -I/usr/src/redhat/BUILD/linux/include -Wall -Wstrict-prototypes -
Wno-trigraphs -O2 -fomit-frame-pointer -fno-strict-aliasing -D__powerpc__ -
fsigned-char -msoft-float -pipe -ffixed-r2 -Wno-uninitialized -mmultiple -mstring
  -c -o softirq.o softirq.c
softirq.c: In function `wakeup_softirqd':
softirq.c:58: structure has no member named `__ksoftirqd_task'
softirq.c: In function `ksoftirqd':
softirq.c:393: structure has no member named `__ksoftirqd_task'
softirq.c: In function `spawn_ksoftirqd':
softirq.c:420: structure has no member named `__ksoftirqd_task'
make[2]: *** [softirq.o] Error 1
make[2]: Leaving directory `/usr/src/redhat/BUILD/linux/kernel'
make[1]: *** [first_rule] Error 2
make[1]: Leaving directory `/usr/src/redhat/BUILD/linux/kernel'
make: *** [_dir_kernel] Error 2
error: Bad exit status from /var/tmp/rpm-tmp.33988 (%build)
-----------------------------


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Kernel-2.4.7.pre5

[Olaf Hering]

On Tue, Jul 10, Kaoru Fukui wrote:

>
> Hi!
>
> I had this problem in linuxppc_2_4 bk tree's linux-2.4.7.pre5.
> Is my fix correct ?

The "correct" fix would be somthing like that:
/mirror/kernel/people/andrea/kernels/v2.4/2.4.7pre3aa1/00_ksoftirqd-7_ppc-2

But Paul was/is not very happy about that softirq thing...



Gruss Olaf

--
 $ man clone

BUGS
       Main feature not yet implemented...

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Kernel-2.4.7.pre5

[Jeff Garzik]

Olaf Hering wrote:
>
> On Tue, Jul 10, Kaoru Fukui wrote:
>
> >
> > Hi!
> >
> > I had this problem in linuxppc_2_4 bk tree's linux-2.4.7.pre5.
> > Is my fix correct ?
>
> The "correct" fix would be somthing like that:
> /mirror/kernel/people/andrea/kernels/v2.4/2.4.7pre3aa1/00_ksoftirqd-7_ppc-2
>
> But Paul was/is not very happy about that softirq thing...

*shrug*   linux-2.4.7-pre5 softirqs should finally (a) have a coherent
implementation with no special cases, and (b) sort through all the bugs
and API weirdness that Ingo added in 2.4.6-preXX.

So Paul can be not happy but Linus is happy with the API, which is the
more important thing...

--
Jeff Garzik      | A recent study has shown that too much soup
Building 1024    | can cause malaise in laboratory mice.
MandrakeSoft     |

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Kernel-2.4.7.pre5

[Tom Rini]

On Tue, Jul 10, 2001 at 02:21:13PM +0200, Olaf Hering wrote:
>
> On Tue, Jul 10, Kaoru Fukui wrote:
>
> >
> > Hi!
> >
> > I had this problem in linuxppc_2_4 bk tree's linux-2.4.7.pre5.
> > Is my fix correct ?
>
> The "correct" fix would be somthing like that:
> /mirror/kernel/people/andrea/kernels/v2.4/2.4.7pre3aa1/00_ksoftirqd-7_ppc-2
>
> But Paul was/is not very happy about that softirq thing...

iirc, Paul doesn't think the first chunk is still needed. The 2nd chunk is
needed for pre4/pre5, but the last chunk looks wierd.  It removes
__cpu_raise_softirq on PPC, which in pre5 all arches still have (x86 is
slightly different then the rest).

--
Tom Rini (TR1265)
http://gate.crashing.org/~trini/

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Kernel-2.4.7.pre5

[Olaf Hering]

On Tue, Jul 10, Tom Rini wrote:

> On Tue, Jul 10, 2001 at 02:21:13PM +0200, Olaf Hering wrote:
> > /mirror/kernel/people/andrea/kernels/v2.4/2.4.7pre3aa1/00_ksoftirqd-7_ppc-2
                                                     ^^^
> > But Paul was/is not very happy about that softirq thing...
>
> iirc, Paul doesn't think the first chunk is still needed. The 2nd chunk is
> needed for pre4/pre5, but the last chunk looks wierd.  It removes
> __cpu_raise_softirq on PPC, which in pre5 all arches still have (x86 is
> slightly different then the rest).

note the pre version string ;)
Just tweak it a bit.


Gruss Olaf

--
 $ man clone

BUGS
       Main feature not yet implemented...

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Kernel-2.4.7.pre5

[Paul Mackerras]

Jeff Garzik writes:

> > The "correct" fix would be somthing like that:
> > /mirror/kernel/people/andrea/kernels/v2.4/2.4.7pre3aa1/00_ksoftirqd-7_ppc-2
> >
> > But Paul was/is not very happy about that softirq thing...
>
> *shrug*   linux-2.4.7-pre5 softirqs should finally (a) have a coherent
> implementation with no special cases, and (b) sort through all the bugs
> and API weirdness that Ingo added in 2.4.6-preXX.

I just looked and the 2.4.7-pre6 implementation looks OK to me.  We do
need the __ksoftirq_task field in the irq_cpustat_t structure.

I remain unconvinced by some of the arguments for having the ksoftirqd
thread(s).  In particular the notion that you reduce your maximum
stack usage that way is bogus - once the softirq is marked pending it
will be executed the next time we take an interrupt, which means that
it can be executed at any time that interrupts are enabled.  But I
don't think the current implementation is wrong, just a bit
unnecessarily complex.  As Jeff says, *shrug*.

Paul.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/