Skip to contentSkip to conversation

[PATCH} x86_64 PM_TRACE support.

PATCH Nigel Cunningham 6 participants 8 messages May 26, 2007 – May 31, 2007

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

Hi all.

As promised I took another look at the patch and at what Randy had prepared to fix the IA64 compilation error. I did some more work on it, and believe that the following is the tidiest correct solution I can come up with. It differs from the version that caused the compilation error primarily in that:

* the #include <asm/resume-trace.h> is inside the #ifdef CONFIG_PM_TRACE.
* now-unnecessary protection for multiple #includes and ifdef testing of CONFIG_PM_TRACE in the asm code were removed.
* do-nothing definitions for !PM_TRACE restored to include/linux/resume-trace.h.

We're therefore depending upon kernel/power/Kconfig having the right depends condition. As far as I can see, IA64 doesn't define CONFIG_X86. Is that correct, or do we need to have (X86 && !IA64)?

Randy, since this is substantially different to the work you did, I haven't added your signed-off-by as we previously discussed.

 arch/x86_64/kernel/vmlinux.lds.S  |    7 +++++++
 drivers/base/power/trace.c        |    5 ++++-
 include/asm-i386/resume-trace.h   |   13 +++++++++++++
 include/asm-x86_64/resume-trace.h |   13 +++++++++++++
 include/linux/resume-trace.h      |   19 +++++--------------
 kernel/power/Kconfig              |    2 +-
 6 files changed, 43 insertions(+), 16 deletions(-)
arch/x86_64/kernel/vmlinux.lds.S +7 -0
diff -ruNp 200.patch-old/arch/x86_64/kernel/vmlinux.lds.S 200.patch-new/arch/x86_64/kernel/vmlinux.lds.S
@@ -52,6 +52,13 @@ SECTIONS
 
   RODATA
 
+  . = ALIGN(4);
+  .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) {
+  	__tracedata_start = .;
+	*(.tracedata)
+  	__tracedata_end = .;
+  }
+
   . = ALIGN(PAGE_SIZE);        /* Align data segment to page size boundary */
 				/* Data */
   .data : AT(ADDR(.data) - LOAD_OFFSET) {
drivers/base/power/trace.c +4 -1
diff -ruNp 200.patch-old/drivers/base/power/trace.c 200.patch-new/drivers/base/power/trace.c
@@ -142,6 +142,7 @@ void set_trace_device(struct device *dev
 {
 	dev_hash_value = hash_string(DEVSEED, dev->bus_id, DEVHASH);
 }
+EXPORT_SYMBOL(set_trace_device);
 
 /*
  * We could just take the "tracedata" index into the .tracedata
@@ -162,6 +163,7 @@ void generate_resume_trace(void *traceda
 	file_hash_value = hash_string(lineno, file, FILEHASH);
 	set_magic_time(user_hash_value, file_hash_value, dev_hash_value);
 }
+EXPORT_SYMBOL(generate_resume_trace);
 
 extern char __tracedata_start, __tracedata_end;
 static int show_file_hash(unsigned int value)
@@ -170,7 +172,8 @@ static int show_file_hash(unsigned int v
 	char *tracedata;
 
 	match = 0;
-	for (tracedata = &__tracedata_start ; tracedata < &__tracedata_end ; tracedata += 6) {
+	for (tracedata = &__tracedata_start ; tracedata < &__tracedata_end ;
+			tracedata += 2 + sizeof(unsigned long)) {
 		unsigned short lineno = *(unsigned short *)tracedata;
 		const char *file = *(const char **)(tracedata + 2);
 		unsigned int hash = hash_string(lineno, file, FILEHASH);
include/asm-i386/resume-trace.h +13 -0
diff -ruNp 200.patch-old/include/asm-i386/resume-trace.h 200.patch-new/include/asm-i386/resume-trace.h
@@ -0,0 +1,13 @@
+#define TRACE_RESUME(user) do {					\
+	if (pm_trace_enabled) {					\
+		void *tracedata;				\
+		asm volatile("movl $1f,%0\n"			\
+			".section .tracedata,\"a\"\n"		\
+			"1:\t.word %c1\n"			\
+			"\t.long %c2\n"				\
+			".previous"				\
+			:"=r" (tracedata)			\
+			: "i" (__LINE__), "i" (__FILE__));	\
+		generate_resume_trace(tracedata, user);		\
+	}							\
+} while (0)
include/asm-x86_64/resume-trace.h +13 -0
diff -ruNp 200.patch-old/include/asm-x86_64/resume-trace.h 200.patch-new/include/asm-x86_64/resume-trace.h
@@ -0,0 +1,13 @@
+#define TRACE_RESUME(user) do {					\
+	if (pm_trace_enabled) {					\
+		void *tracedata;				\
+		asm volatile("movq $1f,%0\n"			\
+			".section .tracedata,\"a\"\n"		\
+			"1:\t.word %c1\n"			\
+			"\t.quad %c2\n"				\
+			".previous"				\
+			:"=r" (tracedata)			\
+			: "i" (__LINE__), "i" (__FILE__));	\
+		generate_resume_trace(tracedata, user);		\
+	}							\
+} while (0)
include/linux/resume-trace.h +5 -14
diff -ruNp 200.patch-old/include/linux/resume-trace.h 200.patch-new/include/linux/resume-trace.h
@@ -2,6 +2,7 @@
 #define RESUME_TRACE_H
 
 #ifdef CONFIG_PM_TRACE
+#include <asm/resume-trace.h>
 
 extern int pm_trace_enabled;
 
@@ -9,20 +10,10 @@ struct device;
 extern void set_trace_device(struct device *);
 extern void generate_resume_trace(void *tracedata, unsigned int user);
 
-#define TRACE_DEVICE(dev) set_trace_device(dev)
-#define TRACE_RESUME(user) do {					\
-	if (pm_trace_enabled) {					\
-		void *tracedata;				\
-		asm volatile("movl $1f,%0\n"			\
-			".section .tracedata,\"a\"\n"		\
-			"1:\t.word %c1\n"			\
-			"\t.long %c2\n"				\
-			".previous"				\
-			:"=r" (tracedata)			\
-			: "i" (__LINE__), "i" (__FILE__));	\
-		generate_resume_trace(tracedata, user);		\
-	}							\
-} while (0)
+#define TRACE_DEVICE(dev) do { \
+	if (pm_trace_enabled) \
+		set_trace_device(dev); \
+	} while(0)
 
 #else
 
kernel/power/Kconfig +1 -1
diff -ruNp 200.patch-old/kernel/power/Kconfig 200.patch-new/kernel/power/Kconfig
@@ -50,7 +50,7 @@ config DISABLE_CONSOLE_SUSPEND
 
 config PM_TRACE
 	bool "Suspend/resume event tracing"
-	depends on PM && PM_DEBUG && X86_32 && EXPERIMENTAL
+	depends on PM && PM_DEBUG && X86 && EXPERIMENTAL
 	default n
 	---help---
 	This enables some cheesy code to save the last PM event point in the
#1 lore raw

Hi!

Quoted 13 lines

As promised I took another look at the patch and at what Randy had prepared to fix the IA64 compilation error. I did some more work on it, and believe that the following is the tidiest correct solution I can come up with. It differs from the version that caused the compilation error primarily in that:

* the #include <asm/resume-trace.h> is inside the #ifdef CONFIG_PM_TRACE.
* now-unnecessary protection for multiple #includes and ifdef testing of CONFIG_PM_TRACE in the asm code were removed.
* do-nothing definitions for !PM_TRACE restored to include/linux/resume-trace.h.

Quoted 3 lines

We're therefore depending upon kernel/power/Kconfig having the right depends condition. As far as I can see, IA64 doesn't define CONFIG_X86. Is that correct, or do we need to have (X86 && !IA64)?

ia64? did you mean x86-64?

Otherwise looks ok to me.

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
#2 lore raw

Pavel Machek pisze:

Quoted 15 lines

Hi!

Quoted 13 lines

As promised I took another look at the patch and at what Randy had prepared to fix the IA64 compilation error. I did some more work on it, and believe that the following is the tidiest correct solution I can come up with. It differs from the version that caused the compilation error primarily in that:

* the #include <asm/resume-trace.h> is inside the #ifdef CONFIG_PM_TRACE.
* now-unnecessary protection for multiple #includes and ifdef testing of CONFIG_PM_TRACE in the asm code were removed.
* do-nothing definitions for !PM_TRACE restored to include/linux/resume-trace.h.

Quoted 8 lines
Quoted 3 lines

We're therefore depending upon kernel/power/Kconfig having the right depends condition. As far as I can see, IA64 doesn't define CONFIG_X86. Is that correct, or do we need to have (X86 && !IA64)?

ia64? did you mean x86-64?

Otherwise looks ok to me.

IIRC enabling pm_trace on x86_64 was breaking compilation on ia64, so I think Nigel meant the latter.

#3 lore raw

Hi.

Quoted 20 lines · Julian Sikorski

Pavel Machek pisze:

Quoted 19 lines

Hi!

Quoted 16 lines

As promised I took another look at the patch and at what Randy had prepared to fix the IA64 compilation error. I did some more work on it, and believe that the following is the tidiest correct solution I can come up with. It differs from the version that caused the compilation error primarily in that:

* the #include <asm/resume-trace.h> is inside the #ifdef CONFIG_PM_TRACE.
* now-unnecessary protection for multiple #includes and ifdef testing of CONFIG_PM_TRACE in the asm code were removed.
* do-nothing definitions for !PM_TRACE restored to include/linux/resume-trace.h.

We're therefore depending upon kernel/power/Kconfig having the right depends condition. As far as I can see, IA64 doesn't define CONFIG_X86. Is that correct, or do we need to have (X86 && !IA64)?

Quoted 6 lines
Quoted 4 lines

ia64? did you mean x86-64?

Otherwise looks ok to me.

IIRC enabling pm_trace on x86_64 was breaking compilation on ia64, so I think Nigel meant the latter.

Yes, it was breaking ia64, so I meant what I said - I can test x86_64 easily. I was concerned about ensuring the condition was right for ia64.

Regards,

Nigel

#4 lore raw
In reply to Nigel Cunningham
Quoted 15 lines · On Sat, 26 May 2007 16:32:54 +1000 Nigel Cunningham wrote:

Hi all.

As promised I took another look at the patch and at what Randy had prepared to fix the IA64 compilation error. I did some more work on it, and believe that the following is the tidiest correct solution I can come up with. It differs from the version that caused the compilation error primarily in that:

* the #include <asm/resume-trace.h> is inside the #ifdef CONFIG_PM_TRACE.
* now-unnecessary protection for multiple #includes and ifdef testing of CONFIG_PM_TRACE in the asm code were removed.
* do-nothing definitions for !PM_TRACE restored to include/linux/resume-trace.h.

Quoted 3 lines

We're therefore depending upon kernel/power/Kconfig having the right depends condition. As far as I can see, IA64 doesn't define CONFIG_X86. Is that correct, or do we need to have (X86 && !IA64)?

Correct, X86 is i386 and x86_64. Not IA64.

Quoted 2 lines

Randy, since this is substantially different to the work you did, I haven't added your signed-off-by as we previously discussed.

OK.

---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

#5 lore raw
In reply to Nigel Cunningham

On Sat, May 26, 2007 at 04:32:54PM +1000, Nigel Cunningham wrote:

Hi Nigel,

Quoted 13 lines

As promised I took another look at the patch and at what Randy had prepared to fix the IA64 compilation error. I did some more work on it, and believe that the following is the tidiest correct solution I can come up with. It differs from the version that caused the compilation error primarily in that:

* the #include <asm/resume-trace.h> is inside the #ifdef CONFIG_PM_TRACE.
* now-unnecessary protection for multiple #includes and ifdef testing of CONFIG_PM_TRACE in the asm code were removed.
* do-nothing definitions for !PM_TRACE restored to include/linux/resume-trace.h.

Quoted 3 lines

We're therefore depending upon kernel/power/Kconfig having the right depends condition. As far as I can see, IA64 doesn't define CONFIG_X86. Is that correct, or do we need to have (X86 && !IA64)?

Can you post a copy of this that isn't mangled by quoted-printable encoding? Whilst it looks fine in my MUA, the diff ends up looking like..

diff -ruNp 200.patch-old/arch/x86_64/kernel/vmlinux.lds.S 200.patch-new/arc= h/x86_64/kernel/vmlinux.lds.S
--- 200.patch-old/arch/x86_64/kernel/vmlinux.lds.S 2007-05-26 14:18:22.0000=
00000 +1000
+++ 200.patch-new/arch/x86_64/kernel/vmlinux.lds.S 2007-05-26 14:23:52.0000=
00000 +1000
@@ -52,6 +52,13 @@ SECTIONS
=20
RODATA
=20
+ . =3D ALIGN(4);
+ .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) {

+       __tracedata_start =3D .;
+       *(.tracedata)
+       __tracedata_end =3D .;

+ }

I'm beginning to think we really need a Documentation/Unhorking-MUAs-HOWTO judging by the amount of broken encodings we seem to get to lkml these days.

Dave

Signature · 1 line
http://www.codemonkey.org.uk
#6 lore raw
Quoted 17 lines · On Wed, 30 May 2007 20:01:34 -0400 Dave Jones wrote:

On Sat, May 26, 2007 at 04:32:54PM +1000, Nigel Cunningham wrote:

Hi Nigel,

Quoted 13 lines

As promised I took another look at the patch and at what Randy had prepared to fix the IA64 compilation error. I did some more work on it, and believe that the following is the tidiest correct solution I can come up with. It differs from the version that caused the compilation error primarily in that:

* the #include <asm/resume-trace.h> is inside the #ifdef CONFIG_PM_TRACE.
* now-unnecessary protection for multiple #includes and ifdef testing of CONFIG_PM_TRACE in the asm code were removed.
* do-nothing definitions for !PM_TRACE restored to include/linux/resume-trace.h.

Quoted 6 lines
Quoted 3 lines

We're therefore depending upon kernel/power/Kconfig having the right depends condition. As far as I can see, IA64 doesn't define CONFIG_X86. Is that correct, or do we need to have (X86 && !IA64)?

Can you post a copy of this that isn't mangled by quoted-printable encoding? Whilst it looks fine in my MUA, the diff ends up looking like..

yep :(

http://www.xenotime.net/linux/patches/x8664-pm-trace-support.patch

in case Nigel is asleep.

Quoted 2 lines

I'm beginning to think we really need a Documentation/Unhorking-MUAs-HOWTO judging by the amount of broken encodings we seem to get to lkml these days.

---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

#7 lore raw

Hi.

Quoted 21 lines · Randy Dunlap
Quoted 21 lines · Nigel Cunningham

Hi Nigel,

Quoted 16 lines

As promised I took another look at the patch and at what Randy had prepared to fix the IA64 compilation error. I did some more work on it, and believe that the following is the tidiest correct solution I can come up with. It differs from the version that caused the compilation error primarily in that:

* the #include <asm/resume-trace.h> is inside the #ifdef CONFIG_PM_TRACE.
* now-unnecessary protection for multiple #includes and ifdef testing of CONFIG_PM_TRACE in the asm code were removed.
* do-nothing definitions for !PM_TRACE restored to include/linux/resume-trace.h.

We're therefore depending upon kernel/power/Kconfig having the right depends condition. As far as I can see, IA64 doesn't define CONFIG_X86. Is that correct, or do we need to have (X86 && !IA64)?

Quoted 8 lines
Quoted 2 lines

Can you post a copy of this that isn't mangled by quoted-printable encoding? Whilst it looks fine in my MUA, the diff ends up looking like..

yep :(

http://www.xenotime.net/linux/patches/x8664-pm-trace-support.patch

in case Nigel is asleep.

Thanks Randy. I wasn't asleep, but was out for most of the day.

Quoted 2 lines
Quoted 2 lines

I'm beginning to think we really need a Documentation/Unhorking-MUAs-HOWTO judging by the amount of broken encodings we seem to get to lkml these days.

That would be good. I just took a look through evo's settings, but can't find whatever option is set wrong. I think it's devolved on me :\.

Regards,

Nigel