RTC support on ia64

Linux-ia64 Joel GUILLET 7 participants 12 messages Dec 17, 2002 – Jul 2, 2003

Share & tools

Joel GUILLET

author

2002-12-17 16:26 UTC #0 lore raw

Hi,

Here is a patch to use the Real Time Clock (RTC) on ia64.
It works on a tiger. It doesn't on early lion machines. (Why ?) I don't know about others ia64 machines. Please report me - if you test it.
To see if it works, please check the /var/log/messages. The driver reports sometimes that "rtc losts some interrupts..." It usually means that the rtc doesn't work, and the interruptions that arrived were coming from a software-timer (see the driver code).

This patch is very simple.
I changed the irq affected for the rtc as it is an "ex-i386" interruption. (an Ox20 offset)
And copied some files from the asm-i386 to asm-ia64...
The duplication of these files is a waste of disk space, but it's coherent with the way it has been made on i386.

You may want to try this patch and test it.
There is a program test included in the Documentation/rtc.txt of the kernel tree.

The actual priority of the interruption is low but perhaps it might be changed (Do anyone know how to make it ?) for some usages.

For the kernel releases after 2.5.45, I've seen that the Kconfig will be modified so that the option will not be available (unless you activate it manually).

This can be used for example to run scheduler latencies tests (a way of using amlat) - I've got some more stuff for this, please mail me.

Here is the patch :

********************************************

arch/ia64/kernel/ia64_ksyms.c +2 -0

diff -urN linux-2.5.45/arch/ia64/kernel/ia64_ksyms.c linux-2.5.45-clean/arch/ia64/kernel/ia64_ksyms.c
@@ -143,3 +143,5 @@
 #endif
 EXPORT_SYMBOL(machvec_noop);
 
+extern spinlock_t rtc_lock;
+EXPORT_SYMBOL(rtc_lock);

arch/ia64/kernel/time.c +1 -0

diff -urN linux-2.5.45/arch/ia64/kernel/time.c linux-2.5.45-clean/arch/ia64/kernel/time.c
@@ -27,6 +27,7 @@
 extern rwlock_t xtime_lock;
 extern unsigned long wall_jiffies;
 extern unsigned long last_time_offset;
+spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED;
 
 u64 jiffies_64;

include/asm-ia64/mc146818rtc.h +21 -2

diff -urN linux-2.5.45/include/asm-ia64/mc146818rtc.h linux-2.5.45-clean/include/asm-ia64/mc146818rtc.h
@@ -1,10 +1,29 @@
+/*
+ * Machine dependent access functions for RTC registers.
+ */
 #ifndef _ASM_IA64_MC146818RTC_H
 #define _ASM_IA64_MC146818RTC_H
 
+#include <asm/io.h>
+
+#ifndef RTC_PORT
+#define RTC_PORT(x)	(0x70 + (x))
+#define RTC_ALWAYS_BCD	1	/* RTC operates in binary mode */
+#endif
+
 /*
- * Machine dependent access functions for RTC registers.
+ * The yet supported machines all access the RTC index register via
+ * an ISA port access but the way to access the date register differs ...
  */
+#define CMOS_READ(addr) ({ \
+outb_p((addr),RTC_PORT(0)); \
+inb_p(RTC_PORT(1)); \
+})
+#define CMOS_WRITE(val, addr) ({ \
+outb_p((addr),RTC_PORT(0)); \
+outb_p((val),RTC_PORT(1)); \
+})
 
-/* empty include file to satisfy the include in genrtc.c */
+#define RTC_IRQ 40 /* irq on a i386 + Ox20  */
 
 #endif /* _ASM_IA64_MC146818RTC_H */

include/asm-ia64/rtc.h +10 -0

diff -urN linux-2.5.45/include/asm-ia64/rtc.h linux-2.5.45-clean/include/asm-ia64/rtc.h
@@ -0,0 +1,10 @@
+#ifndef _IA64_RTC_H
+#define _IA64_RTC_H
+
+/*
+ * ia64 uses the default access methods for the RTC.
+ */
+
+#include <asm-generic/rtc.h>
+
+#endif

**************************************************************

--------------
** Joel

Alex Williamson

2002-12-17 20:13 UTC #1 lore raw

Joel,

Does this provide more functionality than the EFI RTC?
Concerns I have w/ it on ia64 is that the rtc driver assumes you have a legacy rtc at the legacy irq and port address. This won't work on HP zx1 boxes, the legacy hardware doesn't exist. In the future, there's a possibility that the port address could be allocated to non-legacy purposes. Maybe ACPI could tell you if you have a legacy RTC, it could at least tell you if you support PC/AT compatible interrupts. Thanks,

Alex

Quoted 107 lines · Joel GUILLET

Hi,
Here is a patch to use the Real Time Clock (RTC) on ia64.
It works on a tiger. It doesn't on early lion machines. (Why ?) I don't know about others ia64 machines. Please report me - if you test it.
To see if it works, please check the /var/log/messages. The driver reports sometimes that "rtc losts some interrupts..." It usually means that the rtc doesn't work, and the interruptions that arrived were coming from a software-timer (see the driver code).
This patch is very simple.
I changed the irq affected for the rtc as it is an "ex-i386" interruption. (an Ox20 offset)
And copied some files from the asm-i386 to asm-ia64...
The duplication of these files is a waste of disk space, but it's coherent with the way it has been made on i386.
You may want to try this patch and test it.
There is a program test included in the Documentation/rtc.txt of the kernel tree.
The actual priority of the interruption is low but perhaps it might be changed (Do anyone know how to make it ?) for some usages.
For the kernel releases after 2.5.45, I've seen that the Kconfig will be modified so that the option will not be available (unless you activate it manually).
This can be used for example to run scheduler latencies tests (a way of using amlat) - I've got some more stuff for this, please mail me.
Here is the patch :
********************************************
diff -urN linux-2.5.45/arch/ia64/kernel/ia64_ksyms.c linux-2.5.45-clean/arch/ia64/kernel/ia64_ksyms.c
arch/ia64/kernel/ia64_ksyms.c Tue Dec 17 15:08:19 2002 renamed from arch/ia64/kernel/ia64_ksyms.c Thu Oct 31 01:42:54 2002 +2 -0
@@ -143,3 +143,5 @@
 #endif
 EXPORT_SYMBOL(machvec_noop);
 
+extern spinlock_t rtc_lock;
+EXPORT_SYMBOL(rtc_lock);
diff -urN linux-2.5.45/arch/ia64/kernel/time.c linux-2.5.45-clean/arch/ia64/kernel/time.c
arch/ia64/kernel/time.c Tue Dec 17 15:08:19 2002 renamed from arch/ia64/kernel/time.c Thu Oct 31 01:42:20 2002 +1 -0
@@ -27,6 +27,7 @@
 extern rwlock_t xtime_lock;
 extern unsigned long wall_jiffies;
 extern unsigned long last_time_offset;
+spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED;
 
 u64 jiffies_64;
 
diff -urN linux-2.5.45/include/asm-ia64/mc146818rtc.h linux-2.5.45-clean/include/asm-ia64/mc146818rtc.h
include/asm-ia64/mc146818rtc.h Tue Dec 17 15:10:12 2002 renamed from include/asm-ia64/mc146818rtc.h Thu Oct 31 01:43:38 2002 +21 -2
@@ -1,10 +1,29 @@
+/*
+ * Machine dependent access functions for RTC registers.
+ */
 #ifndef _ASM_IA64_MC146818RTC_H
 #define _ASM_IA64_MC146818RTC_H
 
+#include <asm/io.h>
+
+#ifndef RTC_PORT
+#define RTC_PORT(x)    (0x70 + (x))
+#define RTC_ALWAYS_BCD 1       /* RTC operates in binary mode */
+#endif
+
 /*
- * Machine dependent access functions for RTC registers.
+ * The yet supported machines all access the RTC index register via
+ * an ISA port access but the way to access the date register differs ...
  */
+#define CMOS_READ(addr) ({ \
+outb_p((addr),RTC_PORT(0)); \
+inb_p(RTC_PORT(1)); \
+})
+#define CMOS_WRITE(val, addr) ({ \
+outb_p((addr),RTC_PORT(0)); \
+outb_p((val),RTC_PORT(1)); \
+})
 
-/* empty include file to satisfy the include in genrtc.c */
+#define RTC_IRQ 40 /* irq on a i386 + Ox20  */
 
 #endif /* _ASM_IA64_MC146818RTC_H */
diff -urN linux-2.5.45/include/asm-ia64/rtc.h linux-2.5.45-clean/include/asm-ia64/rtc.h
include/asm-ia64/rtc.h Tue Dec 17 15:09:03 2002 renamed from include/asm-ia64/rtc.h Thu Jan 1 01:00:00 1970 +10 -0
@@ -0,0 +1,10 @@
+#ifndef _IA64_RTC_H
+#define _IA64_RTC_H
+
+/*
+ * ia64 uses the default access methods for the RTC.
+ */
+
+#include <asm-generic/rtc.h>
+
+#endif
**************************************************************

Quoted 7 lines

--------------
** Joel
_______________________________________________
Linux-IA64 mailing list
[email protected]
http://lists.linuxia64.org/lists/listinfo/linux-ia64

--
Alex Williamson Linux Development Lab
[email protected] Hewlett Packard
970-898-9173 Fort Collins, CO

Peter Chubb

2002-12-17 21:57 UTC #2 lore raw

In reply to Joel GUILLET

Quoted 1 line

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Alex" = Alex Williamson <[email protected]> writes:

Alex> Joel, Does this provide more functionality than the EFI RTC? Alex> Concerns I have w/ it on ia64 is that the rtc driver assumes you Alex> have a legacy rtc at the legacy irq and port address. This Alex> won't work on HP zx1 boxes, the legacy hardware doesn't exist. Alex> In the future, there's a possibility that the port address could Alex> be allocated to non-legacy purposes. Maybe ACPI could tell you Alex> if you have a legacy RTC, it could at least tell you if you Alex> support PC/AT compatible interrupts. Thanks,

Yes it does. There's no way to get a regular interrupt into user space from the EFI RTC.

I've been using the PMU to generate regular interrupts, and have a locally-modified amlat benchmark that does that.

--
Dr Peter Chubb [email protected]
You are lost in a maze of BitKeeper repositories, all almost the same.

David Mosberger

2002-12-17 23:57 UTC #3 lore raw

In reply to Joel GUILLET

Quoted 1 line

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line · Joel GUILLET

Joel> This can be used for example to run scheduler latencies tests Joel> (a way of using amlat) - I've got some more stuff for this, Joel> please mail me.

What's wrong with using the PMU for this? That's something that works on all ia64 platforms and doesn't depend on legacy hardware.

	--david

Joel GUILLET

author

2002-12-18 15:01 UTC #4 lore raw

In reply to Joel GUILLET

Quoted 11 lines · Peter Chubb

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Alex" = Alex Williamson <[email protected]> writes:
Alex> Joel, Does this provide more functionality than the EFI RTC? Alex> Concerns I have w/ it on ia64 is that the rtc driver assumes you Alex> have a legacy rtc at the legacy irq and port address. This Alex> won't work on HP zx1 boxes, the legacy hardware doesn't exist. Alex> In the future, there's a possibility that the port address could Alex> be allocated to non-legacy purposes. Maybe ACPI could tell you Alex> if you have a legacy RTC, it could at least tell you if you Alex> support PC/AT compatible interrupts. Thanks,

Peter> Yes it does. There's no way to get a regular interrupt into user Peter> space from the EFI RTC.

The RTC legacy hardware seems to be on the i870 chipset. And if you don't have this one, you may not be able to use it.

Peter> I've been using the PMU to generate regular interrupts, and have a Peter> locally-modified amlat benchmark that does that.

Did you modify anything inside the kernel to make use of this timer ? With PMU, you mean "Performance Monitoring Unit" or "Power Management Utility" ? -- this question may look strange but with the acutal high resolution timer his ACPI PowerManagement timer, I've finally got some doubts.

I had a looked at the PMU code but I couldn't find anything about the way of using this timer ? Is it built-in somewhere ?
Do I need to implement some code in the kernel to use it ? I've been trying to find some docs in the Web for quite a long time ... but without any results ! Does an user interface already exist ?

Thanks for your help.

Regards,

-------------
** Joel

Stephane Eranian

2002-12-18 18:18 UTC #5 lore raw

In reply to Joel GUILLET

Joel,

Quoted 30 lines · Joel GUILLET

Quoted 11 lines · Peter Chubb
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Alex" = Alex Williamson <[email protected]> writes:
Alex> Joel, Does this provide more functionality than the EFI RTC? Alex> Concerns I have w/ it on ia64 is that the rtc driver assumes you Alex> have a legacy rtc at the legacy irq and port address. This Alex> won't work on HP zx1 boxes, the legacy hardware doesn't exist. Alex> In the future, there's a possibility that the port address could Alex> be allocated to non-legacy purposes. Maybe ACPI could tell you Alex> if you have a legacy RTC, it could at least tell you if you Alex> support PC/AT compatible interrupts. Thanks,
Peter> Yes it does. There's no way to get a regular interrupt into user Peter> space from the EFI RTC.
The RTC legacy hardware seems to be on the i870 chipset. And if you don't have this one, you may not be able to use it.
Peter> I've been using the PMU to generate regular interrupts, and have a Peter> locally-modified amlat benchmark that does that.
Did you modify anything inside the kernel to make use of this timer ? With PMU, you mean "Performance Monitoring Unit" or "Power Management Utility" ? -- this question may look strange but with the acutal high resolution timer his ACPI PowerManagement timer, I've finally got some doubts.

Quoted 6 lines

I had a looked at the PMU code but I couldn't find anything about the way of using this timer ? Is it built-in somewhere ?
Do I need to implement some code in the kernel to use it ? I've been trying to find some docs in the Web for quite a long time ... but without any results ! Does an user interface already exist ?

PMU means Performance Monitoring Unit. What David meant is that you can get fine grain timer with the PMU by using the CPU_CYCLES event. You can program a PMU counter to overflow after a certain number of cycles have elapsed. It is very fine grain because you get down to the cycle. If you setup the counter to 2^64-n, then it will overflow after n cycles. Setting this up is relatively easy. I encourage you to download the libpfm library. It contains examples on how to setup counters and get overflow notifications. You can get the library and also the pfmon monitoring tool from:

		ftp://ftp.hpl.hp.com/pub/linux-ia64/libpfm-2.0.tar.gz
		ftp://ftp.hpl.hp.com/pub/linux-ia64/pfmon-2.0.tar.gz

These two package are beta versions of the upcoming 2.0 release.

Signature · 1 line

-Stephane

Alex Williamson

2002-12-18 19:26 UTC #6 lore raw

In reply to Joel GUILLET

Quoted 3 lines · Joel GUILLET

The RTC legacy hardware seems to be on the i870 chipset. And if you don't have this one, you may not be able to use it.

Obviously, but that wasn't really my point. What I'm really
after is that on ia64 you really shouldn't be assuming legacy hardware is present and blindly poking port space and setting up irqs based on that assumption. Not only might it not work on a legacy free box, it might break other things. I think you really need to make some attempt to determine if the hardware is out there, and I don't see one in the rtc_init path.

Alex

--
Alex Williamson Linux Development Lab
[email protected] Hewlett Packard
970-898-9173 Fort Collins, CO

Zhu, Yi

2003-07-01 06:11 UTC #7 lore raw

In reply to Joel GUILLET

Hi Joel,

Do you have the ia64 rtc patch for 2.4 kernel? I need the rtc driver to test the latency with realfeel.

Thanks,
-yi

Quoted 137 lines

-----Original Message-----
From: Joel GUILLET [mailto:[email protected]]
Sent: Wednesday, December 18, 2002 12:26 AM
To: [email protected]
Subject: [Linux-ia64] RTC support on ia64
Hi,
Here is a patch to use the Real Time Clock (RTC) on ia64. It works on a tiger. It doesn't on early lion machines. (Why ?) I don't know about others ia64 machines. Please report me - if you test
it.
To see if it works, please check the /var/log/messages. The driver reports
sometimes that "rtc losts some interrupts..." It usually means that the
rtc doesn't work, and the interruptions that arrived were coming from a
software-timer (see the driver code).
This patch is very simple.
I changed the irq affected for the rtc as it is an "ex-i386" interruption.
(an Ox20 offset)
And copied some files from the asm-i386 to asm-ia64... The duplication of these files is a waste of disk space, but it's coherent
with the way it has been made on i386.
You may want to try this patch and test it.
There is a program test included in the Documentation/rtc.txt of the kernel tree.
The actual priority of the interruption is low but perhaps it might be changed (Do anyone know how to make it ?) for some usages.
For the kernel releases after 2.5.45, I've seen that the Kconfig will be
modified so that the option will not be available (unless you activate it
manually).
This can be used for example to run scheduler latencies tests (a way of
using amlat) - I've got some more stuff for this, please mail me.
Here is the patch :
********************************************
diff -urN linux-2.5.45/arch/ia64/kernel/ia64_ksyms.c linux-2.5.45-clean/arch/ia64/kernel/ia64_ksyms.c
--- linux-2.5.45/arch/ia64/kernel/ia64_ksyms.c Thu Oct
31 01:42:54 2002
+++ linux-2.5.45-clean/arch/ia64/kernel/ia64_ksyms.c Tue Dec
17 15:08:19 2002
arch/ia64/kernel/ia64_ksyms.c +2 -0
@@ -143,3 +143,5 @@
 #endif
 EXPORT_SYMBOL(machvec_noop);
 
+extern spinlock_t rtc_lock;
+EXPORT_SYMBOL(rtc_lock);
diff -urN linux-2.5.45/arch/ia64/kernel/time.c
linux-2.5.45-clean/arch/ia64/kernel/time.c
arch/ia64/kernel/time.c +0 -0
17 15:08:19 2002
arch/ia64/kernel/time.c +1 -0
@@ -27,6 +27,7 @@
 extern rwlock_t xtime_lock;
 extern unsigned long wall_jiffies;
 extern unsigned long last_time_offset;
+spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED;
 
 u64 jiffies_64;
 
diff -urN linux-2.5.45/include/asm-ia64/mc146818rtc.h linux-2.5.45-clean/include/asm-ia64/mc146818rtc.h
--- linux-2.5.45/include/asm-ia64/mc146818rtc.h Thu Oct
31 01:43:38 2002
+++ linux-2.5.45-clean/include/asm-ia64/mc146818rtc.h Tue Dec
17 15:10:12 2002
arch/ia64/kernel/time.c +8 -0
@@ -1,10 +1,29 @@
+/*
+ * Machine dependent access functions for RTC registers.
+ */
 #ifndef _ASM_IA64_MC146818RTC_H
 #define _ASM_IA64_MC146818RTC_H
 
+#include <asm/io.h>
+
+#ifndef RTC_PORT
+#define RTC_PORT(x)	(0x70 + (x))
+#define RTC_ALWAYS_BCD	1	/* RTC operates in 
binary mode */
arch/ia64/kernel/time.c +4 -1
+#endif
+
 /*
- * Machine dependent access functions for RTC registers.
+ * The yet supported machines all access the RTC index register via
+ * an ISA port access but the way to access the date 
register differs ...
*/
arch/ia64/kernel/time.c +9 -1
+#define CMOS_READ(addr) ({ \
+outb_p((addr),RTC_PORT(0)); \
+inb_p(RTC_PORT(1)); \
+})
+#define CMOS_WRITE(val, addr) ({ \
+outb_p((addr),RTC_PORT(0)); \
+outb_p((val),RTC_PORT(1)); \
+})
 
-/* empty include file to satisfy the include in genrtc.c */
+#define RTC_IRQ 40 /* irq on a i386 + Ox20  */
 
 #endif /* _ASM_IA64_MC146818RTC_H */
diff -urN linux-2.5.45/include/asm-ia64/rtc.h
linux-2.5.45-clean/include/asm-ia64/rtc.h
include/asm-ia64/rtc.h +10 -0
@@ -0,0 +1,10 @@
+#ifndef _IA64_RTC_H
+#define _IA64_RTC_H
+
+/*
+ * ia64 uses the default access methods for the RTC.
+ */
+
+#include <asm-generic/rtc.h>
+
+#endif
**************************************************************

Quoted 9 lines

--------------
** Joel
_______________________________________________
Linux-IA64 mailing list
[email protected]
http://lists.linuxia64.org/lists/listinfo/linux-ia64

_______________________________________________
Linux-IA64 mailing list
[email protected]
http://lists.linuxia64.org/lists/listinfo/linux-ia64

Zhu, Yi

2003-07-01 06:11 UTC #8 lore raw

In reply to Joel GUILLET

Hi Joel,

Do you have the ia64 rtc patch for 2.4 kernel? I need the rtc driver to test the latency with realfeel.

Thanks,
-yi

Quoted 146 lines

-----Original Message-----
From: Joel GUILLET [mailto:[email protected]]
Sent: Wednesday, December 18, 2002 12:26 AM
To: [email protected]
Subject: [Linux-ia64] RTC support on ia64
Hi,
Here is a patch to use the Real Time Clock (RTC) on ia64. It works on a tiger. It doesn't on early lion machines. (Why ?) I don't know about others ia64 machines. Please report me - if you test
it.
To see if it works, please check the /var/log/messages. The driver reports
sometimes that "rtc losts some interrupts..." It usually means that the
rtc doesn't work, and the interruptions that arrived were coming from a
software-timer (see the driver code).
This patch is very simple.
I changed the irq affected for the rtc as it is an "ex-i386" interruption.
(an Ox20 offset)
And copied some files from the asm-i386 to asm-ia64... The duplication of these files is a waste of disk space, but it's coherent
with the way it has been made on i386.
You may want to try this patch and test it.
There is a program test included in the Documentation/rtc.txt of the kernel tree.
The actual priority of the interruption is low but perhaps it might be changed (Do anyone know how to make it ?) for some usages.
For the kernel releases after 2.5.45, I've seen that the Kconfig will be
modified so that the option will not be available (unless you activate it
manually).
This can be used for example to run scheduler latencies tests (a way of
using amlat) - I've got some more stuff for this, please mail me.
Here is the patch :
********************************************
diff -urN linux-2.5.45/arch/ia64/kernel/ia64_ksyms.c linux-2.5.45-clean/arch/ia64/kernel/ia64_ksyms.c
--- linux-2.5.45/arch/ia64/kernel/ia64_ksyms.c Thu Oct
31 01:42:54 2002
+++ linux-2.5.45-clean/arch/ia64/kernel/ia64_ksyms.c Tue Dec
17 15:08:19 2002
arch/ia64/kernel/ia64_ksyms.c +2 -0
@@ -143,3 +143,5 @@
 #endif
 EXPORT_SYMBOL(machvec_noop);
 
+extern spinlock_t rtc_lock;
+EXPORT_SYMBOL(rtc_lock);
diff -urN linux-2.5.45/arch/ia64/kernel/time.c
linux-2.5.45-clean/arch/ia64/kernel/time.c
arch/ia64/kernel/time.c +0 -0
17 15:08:19 2002
arch/ia64/kernel/time.c +1 -0
@@ -27,6 +27,7 @@
 extern rwlock_t xtime_lock;
 extern unsigned long wall_jiffies;
 extern unsigned long last_time_offset;
+spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED;
 
 u64 jiffies_64;
 
diff -urN linux-2.5.45/include/asm-ia64/mc146818rtc.h linux-2.5.45-clean/include/asm-ia64/mc146818rtc.h
--- linux-2.5.45/include/asm-ia64/mc146818rtc.h Thu Oct
31 01:43:38 2002
+++ linux-2.5.45-clean/include/asm-ia64/mc146818rtc.h Tue Dec
17 15:10:12 2002
arch/ia64/kernel/time.c +8 -0
@@ -1,10 +1,29 @@
+/*
+ * Machine dependent access functions for RTC registers.
+ */
 #ifndef _ASM_IA64_MC146818RTC_H
 #define _ASM_IA64_MC146818RTC_H
 
+#include <asm/io.h>
+
+#ifndef RTC_PORT
+#define RTC_PORT(x)	(0x70 + (x))
+#define RTC_ALWAYS_BCD	1	/* RTC operates in 
binary mode */
arch/ia64/kernel/time.c +4 -1
+#endif
+
 /*
- * Machine dependent access functions for RTC registers.
+ * The yet supported machines all access the RTC index register via
+ * an ISA port access but the way to access the date 
register differs ...
*/
arch/ia64/kernel/time.c +9 -1
+#define CMOS_READ(addr) ({ \
+outb_p((addr),RTC_PORT(0)); \
+inb_p(RTC_PORT(1)); \
+})
+#define CMOS_WRITE(val, addr) ({ \
+outb_p((addr),RTC_PORT(0)); \
+outb_p((val),RTC_PORT(1)); \
+})
 
-/* empty include file to satisfy the include in genrtc.c */
+#define RTC_IRQ 40 /* irq on a i386 + Ox20  */
 
 #endif /* _ASM_IA64_MC146818RTC_H */
diff -urN linux-2.5.45/include/asm-ia64/rtc.h
linux-2.5.45-clean/include/asm-ia64/rtc.h
include/asm-ia64/rtc.h +10 -0
@@ -0,0 +1,10 @@
+#ifndef _IA64_RTC_H
+#define _IA64_RTC_H
+
+/*
+ * ia64 uses the default access methods for the RTC.
+ */
+
+#include <asm-generic/rtc.h>
+
+#endif
**************************************************************
--------------
** Joel
_______________________________________________
Linux-IA64 mailing list
[email protected]
http://lists.linuxia64.org/lists/listinfo/linux-ia64

Eric Piel

2003-07-01 12:15 UTC #9 lore raw

In reply to Joel GUILLET

Quoted 5 lines · Yi

Hi Joel,
Do you have the ia64 rtc patch for 2.4 kernel? I need the rtc driver to test the latency with realfeel.

Hello Yi,
Joel was a trainee, so he's not here anymore and I'm the new one.

We do not have any 2.4 version but the backport shouldn't be hard. When Joel did this patch it was also in order to get realfeel working. The point is that it seems on ia64 measuring latencies using the perfmon capabilities of the processor is a much better idea.

Peter Chubb even developed a small realfeel equivalent that does exactly that! It should work both on 2.4 and 2.5 kernels. You can find it on the perfmon homepage and at this address:
http://www.gelato.unsw.edu.au/patches/realfeel4.c

Cheers
Eric

Peter Chubb

2003-07-01 22:53 UTC #10 lore raw

In reply to Joel GUILLET

Quoted 1 line

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Eric" = Eric Piel <[email protected]> writes:

Eric> Peter Chubb even developed a small realfeel equivalent that does Eric> exactly that! It should work both on 2.4 and 2.5 kernels. You Eric> can find it on the perfmon homepage and at this address:
Eric> http://www.gelato.unsw.edu.au/patches/realfeel4.c

I've also developed a microstate accounting patch that allows you to measure the times in the interrupt handler and on the ACTIVE and EXPIRED queues directly (and hence gives a direct measurement of latency). The patch was posted to LKML; and there'll be a copy on the gelato web page ( http://www.gelato.unsw.edu.au/patches ) later today.

--
Dr Peter Chubb http://www.gelato.unsw.edu.au peterc AT gelato.unsw.edu.au
You are lost in a maze of BitKeeper repositories, all slightly different.

Zhu, Yi

2003-07-02 00:45 UTC #11 lore raw

In reply to Joel GUILLET

Thanks both! I got realfeel4 work!

-yi

Quoted 35 lines

-----Original Message-----
From: Eric Piel [mailto:[email protected]]
Sent: Tuesday, July 01, 2003 8:15 PM
To: Zhu, Yi
Cc: Joel GUILLET; [email protected]
Subject: Re: [Linux-ia64] RTC support on ia64
Quoted 5 lines · Yi
Hi Joel,
Do you have the ia64 rtc patch for 2.4 kernel? I need the rtc driver to test the latency with realfeel.
Hello Yi,
Joel was a trainee, so he's not here anymore and I'm the new one.
We do not have any 2.4 version but the backport shouldn't be hard. When
Joel did this patch it was also in order to get realfeel working. The point is that it seems on ia64 measuring latencies using the perfmon
capabilities of the processor is a much better idea.
Peter Chubb even developed a small realfeel equivalent that does exactly
that! It should work both on 2.4 and 2.5 kernels. You can find it on the
perfmon homepage and at this address:
http://www.gelato.unsw.edu.au/patches/realfeel4.c
Cheers
Eric