* [Xenomai-core] SMI update
@ 2006-08-09 16:37 Ulrich Schwab
2006-08-09 22:44 ` Jan Kiszka
0 siblings, 1 reply; 6+ messages in thread
From: Ulrich Schwab @ 2006-08-09 16:37 UTC (permalink / raw)
To: xenomai
[-- Attachment #1: Type: text/plain, Size: 724 bytes --]
I saw some bad latency behaviour on a new system, it is a
P4 with ICH7 chipset. It showed more than 300 us max. latency without any
load.
This chipset was not detected as SMI generating and SMI disable was not done.
Finally I figured out where to add this DEVICE_ID
to the SMI detection in smi.c
It is done in the attached patch.
Now, the max. latency values on this new system are below 10 us.
Although, not tested very long.
Regards,
Ulrich Schwab
--
====================================================
inmess GmbH
Frankfurter Str. 74
D - 64521 Gross-Gerau
Phone: +49 6152 97790
Fax : +49 6152 977920
mail : info@domain.hid
web: www.inmess.de
====================================================
[-- Attachment #2: smi-ICH7-patch --]
[-- Type: text/x-diff, Size: 580 bytes --]
diff -urp xenomai-2.2.0/ksrc/arch/i386/smi.c xenomai-2.2-patched/ksrc/arch/i386/smi.c
--- xenomai-2.2.0/ksrc/arch/i386/smi.c 2006-07-03 08:44:33.000000000 +0200
+++ xenomai-2.2-patched/ksrc/arch/i386/smi.c 2006-08-09 18:20:57.000000000 +0200
@@ -45,6 +45,7 @@ static struct pci_device_id rthal_smi_pc
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_0) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_2) },
+{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_0) },
{ 0, },
};
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Xenomai-core] SMI update
2006-08-09 16:37 [Xenomai-core] SMI update Ulrich Schwab
@ 2006-08-09 22:44 ` Jan Kiszka
2006-08-10 7:44 ` Ulrich Schwab
0 siblings, 1 reply; 6+ messages in thread
From: Jan Kiszka @ 2006-08-09 22:44 UTC (permalink / raw)
To: Ulrich Schwab; +Cc: xenomai
[-- Attachment #1: Type: text/plain, Size: 781 bytes --]
Ulrich Schwab wrote:
> I saw some bad latency behaviour on a new system, it is a
> P4 with ICH7 chipset. It showed more than 300 us max. latency without any
> load.
> This chipset was not detected as SMI generating and SMI disable was not done.
>
> Finally I figured out where to add this DEVICE_ID
> to the SMI detection in smi.c
> It is done in the attached patch.
>
> Now, the max. latency values on this new system are below 10 us.
> Although, not tested very long.
...and not under load. ;)
Thanks for the patch, we just need some additional define of
PCI_DEVICE_ID_INTEL_ICH7_0 for kernel 2.4 compatibility. And if you
additionally provide a ChangeLog fragment, your patch would be perfect!
(I tend to forget the last part often too... :) )
Jan
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 250 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Xenomai-core] SMI update
2006-08-09 22:44 ` Jan Kiszka
@ 2006-08-10 7:44 ` Ulrich Schwab
2006-08-10 8:27 ` Jan Kiszka
0 siblings, 1 reply; 6+ messages in thread
From: Ulrich Schwab @ 2006-08-10 7:44 UTC (permalink / raw)
To: xenomai
On Thursday 10 August 2006 00:44, Jan Kiszka wrote:
> Thanks for the patch, we just need some additional define of
> PCI_DEVICE_ID_INTEL_ICH7_0 for kernel 2.4 compatibility. And if you
> additionally provide a ChangeLog fragment, your patch would be perfect!
> (I tend to forget the last part often too... :) )
Where to add the new define for kernel 2.4 ?
Ulrich
--
====================================================
inmess GmbH
Frankfurter Str. 74
D - 64521 Gross-Gerau
Phone: +49 6152 97790
Fax : +49 6152 977920
mail : info@domain.hid
web: www.inmess.de
====================================================
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Xenomai-core] SMI update
2006-08-10 7:44 ` Ulrich Schwab
@ 2006-08-10 8:27 ` Jan Kiszka
2006-08-10 12:54 ` Ulrich Schwab
0 siblings, 1 reply; 6+ messages in thread
From: Jan Kiszka @ 2006-08-10 8:27 UTC (permalink / raw)
To: Ulrich Schwab; +Cc: xenomai
[-- Attachment #1: Type: text/plain, Size: 550 bytes --]
Ulrich Schwab wrote:
> On Thursday 10 August 2006 00:44, Jan Kiszka wrote:
>> Thanks for the patch, we just need some additional define of
>> PCI_DEVICE_ID_INTEL_ICH7_0 for kernel 2.4 compatibility. And if you
>> additionally provide a ChangeLog fragment, your patch would be perfect!
>> (I tend to forget the last part often too... :) )
> Where to add the new define for kernel 2.4 ?
>
I would say to smi.c directly as there are no other users in sight.
Otherwise, wrapping code goes to include/asm-{generic|<arch>}/wrapper.h.
Jan
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 250 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Xenomai-core] SMI update
2006-08-10 8:27 ` Jan Kiszka
@ 2006-08-10 12:54 ` Ulrich Schwab
2006-08-12 20:30 ` Gilles Chanteperdrix
0 siblings, 1 reply; 6+ messages in thread
From: Ulrich Schwab @ 2006-08-10 12:54 UTC (permalink / raw)
To: xenomai
[-- Attachment #1: Type: text/plain, Size: 921 bytes --]
On Thursday 10 August 2006 10:27, Jan Kiszka wrote:
> Ulrich Schwab wrote:
> > On Thursday 10 August 2006 00:44, Jan Kiszka wrote:
> >> Thanks for the patch, we just need some additional define of
> >> PCI_DEVICE_ID_INTEL_ICH7_0 for kernel 2.4 compatibility. And if you
> >> additionally provide a ChangeLog fragment, your patch would be perfect!
> >> (I tend to forget the last part often too... :) )
> >
> > Where to add the new define for kernel 2.4 ?
>
> I would say to smi.c directly as there are no other users in sight.
> Otherwise, wrapping code goes to include/asm-{generic|<arch>}/wrapper.h.
I did add the new Id to smi.c directly now.
Ulrich
--
====================================================
inmess GmbH
Frankfurter Str. 74
D - 64521 Gross-Gerau
Phone: +49 6152 97790
Fax : +49 6152 977920
mail : info@domain.hid
web: www.inmess.de
====================================================
[-- Attachment #2: smi-ICH7-patch-2 --]
[-- Type: text/x-diff, Size: 1495 bytes --]
diff -urp xenomai-2.2.0-org/ChangeLog xenomai-2.2.0-patched/ChangeLog
--- xenomai-2.2.0-org/ChangeLog 2006-07-17 18:50:17.000000000 +0200
+++ xenomai-2.2.0-patched/ChangeLog 2006-08-10 14:51:28.000000000 +0200
@@ -1,3 +1,7 @@
+2006-08-10 Ulrich Schwab <schwab@domain.hid>
+
+ * ksrc/arch/i386/smi.c added id of Intel ICH7 to list of SMI chipsets
+
2006-07-17 Philippe Gerum <rpm@xenomai.org>
* RELEASE: Xenomai 2.2 (Engines Of Creation)
diff -urp xenomai-2.2.0-org/ksrc/arch/i386/smi.c xenomai-2.2.0-patched/ksrc/arch/i386/smi.c
--- xenomai-2.2.0-org/ksrc/arch/i386/smi.c 2006-07-03 08:44:33.000000000 +0200
+++ xenomai-2.2.0-patched/ksrc/arch/i386/smi.c 2006-08-10 14:50:37.000000000 +0200
@@ -31,6 +31,13 @@
#include <linux/reboot.h>
#include <asm/xenomai/smi.h>
+/* DEVICE_IDs needed for 2.4 support */
+#if LINUX_VERSION_CODE < KERNEL_VERSION(2,5,0)
+#ifndef PCI_DEVICE_ID_INTEL_ICH7_0
+#define PCI_DEVICE_ID_INTEL_ICH7_0 0x27b8
+#endif
+#endif
+
static struct pci_device_id rthal_smi_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801AA_0) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801AB_0) },
@@ -45,6 +52,7 @@ static struct pci_device_id rthal_smi_pc
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_0) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_2) },
+{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_0) },
{ 0, },
};
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Xenomai-core] SMI update
2006-08-10 12:54 ` Ulrich Schwab
@ 2006-08-12 20:30 ` Gilles Chanteperdrix
0 siblings, 0 replies; 6+ messages in thread
From: Gilles Chanteperdrix @ 2006-08-12 20:30 UTC (permalink / raw)
To: Ulrich Schwab; +Cc: xenomai
Ulrich Schwab wrote:
> On Thursday 10 August 2006 10:27, Jan Kiszka wrote:
> > Ulrich Schwab wrote:
> > > On Thursday 10 August 2006 00:44, Jan Kiszka wrote:
> > >> Thanks for the patch, we just need some additional define of
> > >> PCI_DEVICE_ID_INTEL_ICH7_0 for kernel 2.4 compatibility. And if you
> > >> additionally provide a ChangeLog fragment, your patch would be perfect!
> > >> (I tend to forget the last part often too... :) )
> > >
> > > Where to add the new define for kernel 2.4 ?
> >
> > I would say to smi.c directly as there are no other users in sight.
> > Otherwise, wrapping code goes to include/asm-{generic|<arch>}/wrapper.h.
> I did add the new Id to smi.c directly now.
Applied, thanks.
--
Gilles Chanteperdrix.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2006-08-12 20:30 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-08-09 16:37 [Xenomai-core] SMI update Ulrich Schwab
2006-08-09 22:44 ` Jan Kiszka
2006-08-10 7:44 ` Ulrich Schwab
2006-08-10 8:27 ` Jan Kiszka
2006-08-10 12:54 ` Ulrich Schwab
2006-08-12 20:30 ` Gilles Chanteperdrix
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.