Skip to contentSkip to conversation

Immedate Values Optimized Jump Fix

PATCH In review Mathieu Desnoyers 3 participants 8 messages Jul 17, 2008 4 messages not in the archive

Subsystems (2)
lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): 20080715024754.GA26471@Krystal

#1 lore raw

I rarely emit a single patch for a fix wrt LTTng (I usually just release a new version), but this one is worth it. It will be in 0.11 pretty soon, but if you want to apply it to 0.10 meanwhile, it's highly recommended.

Caused a kernel OOPS when loading the tracepoint probes once in a while on a 8-way x86_64.

Immedate Values Optimized Jump Fix

Fix the immediate values optimized jump fallback, which parameters were wrong following the last changes. It should be a 5 bytes instruction (not 2) with a 4 bytes operand.

---
 arch/x86/kernel/immediate.c |    2 +-
 include/asm-x86/immediate.h |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
arch/x86/kernel/immediate.c +1 -1
Index: linux-2.6-lttng/arch/x86/kernel/immediate.c
===================================================================
@@ -500,7 +500,7 @@ __kprobes int arch_imv_update(struct __i
 				"Jump target fallback at %lX, nr fail %d\n",
 				imv->imv, ++nr_fail);
 #endif
-			imv->size = 1;
+			imv->size = 4;	/* Fallback on movl */
 		} else {
 #ifdef DEBUG_IMMEDIATE
 			static int nr_success;
include/asm-x86/immediate.h +1 -1
Index: linux-2.6-lttng/include/asm-x86/immediate.h
===================================================================
@@ -132,7 +132,7 @@ struct __imv {
 		BUILD_BUG_ON(sizeof(__typeof__(name##__imv)) > 1);	\
 		asm (".section __imv,\"aw\",@progbits\n\t"		\
 			_ASM_PTR "%c1, (3f)-4\n\t"			\
-			".byte 0, 2\n\t"				\
+			".byte 0, 5\n\t"				\
 			".previous\n\t"					\
 			"mov $0,%0\n\t"					\
 			"3:\n\t"					\
Signature · 2 lines
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68
#2 lore raw

Re: [ltt-dev] LTTng 0.11

LTTng 0.11 is released, which contains this bugfix.

http://ltt.polymtl.ca/lttng/patch-2.6.26-0.11.tar.bz2

Mathieu

Quoted 57 lines · * Mathieu Desnoyers ([email protected]) wrote:

I rarely emit a single patch for a fix wrt LTTng (I usually just release a new version), but this one is worth it. It will be in 0.11 pretty soon, but if you want to apply it to 0.10 meanwhile, it's highly recommended.

Caused a kernel OOPS when loading the tracepoint probes once in a while on a 8-way x86_64.

Immedate Values Optimized Jump Fix

Fix the immediate values optimized jump fallback, which parameters were wrong following the last changes. It should be a 5 bytes instruction (not 2) with a 4 bytes operand.

---
 arch/x86/kernel/immediate.c |    2 +-
 include/asm-x86/immediate.h |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

Index: linux-2.6-lttng/arch/x86/kernel/immediate.c
===================================================================

arch/x86/kernel/immediate.c +1 -1
@@ -500,7 +500,7 @@ __kprobes int arch_imv_update(struct __i
 				"Jump target fallback at %lX, nr fail %d\n",
 				imv->imv, ++nr_fail);
 #endif
-			imv->size = 1;
+			imv->size = 4;	/* Fallback on movl */
 		} else {
 #ifdef DEBUG_IMMEDIATE
 			static int nr_success;

Index: linux-2.6-lttng/include/asm-x86/immediate.h
===================================================================

include/asm-x86/immediate.h +1 -1
@@ -132,7 +132,7 @@ struct __imv {
 		BUILD_BUG_ON(sizeof(__typeof__(name##__imv)) > 1);	\
 		asm (".section __imv,\"aw\",@progbits\n\t"		\
 			_ASM_PTR "%c1, (3f)-4\n\t"			\
-			".byte 0, 2\n\t"				\
+			".byte 0, 5\n\t"				\
 			".previous\n\t"					\
 			"mov $0,%0\n\t"					\
 			"3:\n\t"					\
Signature · 8 lines
-- 
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68

_______________________________________________
ltt-dev mailing list
[email protected]
http://lists.casi.polymtl.ca/cgi-bin/mailman/listinfo/ltt-dev
Signature · 2 lines
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): 20080717024550.GA22869@Krystal

A message that is not in the archive (sent off-list or not yet received): [email protected]

#6 lore raw

PATCH 1/1 Fix markers duplicate modpost entry

When a kernel was rebuilt, the previous Module.markers was not cleared. It caused markers with different format strings to appear as duplicates when a markers was changed.

I merely merged the patches from Roland and Wenji here. It applies to mainline (and is not intrusive, so will also apply to linux-next).

From: Roland McGrath <roland@redhat dot com>
From: Wenji Huang <[email protected]>
CC: [email protected]

---
 scripts/Makefile.modpost |    1 +
 scripts/mod/modpost.c    |    3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)
scripts/Makefile.modpost +1 -0
Index: linux-2.6-lttng/scripts/Makefile.modpost
===================================================================
@@ -101,6 +101,7 @@ quiet_cmd_kernel-mod = MODPOST $@
       cmd_kernel-mod = $(modpost) $@
 
 vmlinux.o: FORCE
+	@rm -fr $(kernelmarkersfile)
 	$(call cmd,kernel-mod)
 
 # Declare generated files as targets for modpost
scripts/mod/modpost.c +2 -1
Index: linux-2.6-lttng/scripts/mod/modpost.c
===================================================================
@@ -1992,7 +1992,8 @@ static void read_markers(const char *fna
 			mod->skip = 1;
 		}
 
-		add_marker(mod, marker, fmt);
+		if (!mod->skip)
+			add_marker(mod, marker, fmt);
 	}
 	return;
 fail:
Signature · 2 lines
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68
#7 lore raw

Re: [PATCH] Fix markers duplicate modpost entry

Quoted 6 lines · On Thu, 17 Jul 2008 03:03:21 -0400 Mathieu Desnoyers…

When a kernel was rebuilt, the previous Module.markers was not cleared. It caused markers with different format strings to appear as duplicates when a markers was changed.

I merely merged the patches from Roland and Wenji here. It applies to mainline (and is not intrusive, so will also apply to linux-next).

Is this fix needed in 2.6.26.x? 2.6.25.x?

Quoted 4 lines

From: Roland McGrath <roland@redhat dot com>
From: Wenji Huang <[email protected]>
CC: [email protected]

whoa, what's all that about?

Patches can only have one Author: in git, and we indicate that in emails by putting a From: line right at the top of the changelog. (Additional credits can and should be mentioned in the changelog text of course). When that From: line is missing we take the authorship info from the email headers.

So according to the above, this patch has three authors. Geeze, even I can type faster than that!

Who wrote it?

Quoted 31 lines
---
 scripts/Makefile.modpost |    1 +
 scripts/mod/modpost.c    |    3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)

Index: linux-2.6-lttng/scripts/Makefile.modpost
===================================================================

scripts/Makefile.modpost +1 -0
@@ -101,6 +101,7 @@ quiet_cmd_kernel-mod = MODPOST $@
       cmd_kernel-mod = $(modpost) $@
 
 vmlinux.o: FORCE
+	@rm -fr $(kernelmarkersfile)
 	$(call cmd,kernel-mod)
 
 # Declare generated files as targets for modpost

Index: linux-2.6-lttng/scripts/mod/modpost.c
===================================================================

scripts/mod/modpost.c +2 -1
@@ -1992,7 +1992,8 @@ static void read_markers(const char *fna
 			mod->skip = 1;
 		}
 
-		add_marker(mod, marker, fmt);
+		if (!mod->skip)
+			add_marker(mod, marker, fmt);
 	}
 	return;
 fail:
#8 lore raw

Re: [PATCH] Fix markers duplicate modpost entry

Quoted 11 lines · * Andrew Morton ([email protected]) wrote:
Quoted 6 lines · On Thu, 17 Jul 2008 03:03:21 -0400 Mathieu Desnoyers…

When a kernel was rebuilt, the previous Module.markers was not cleared. It caused markers with different format strings to appear as duplicates when a markers was changed.

I merely merged the patches from Roland and Wenji here. It applies to mainline (and is not intrusive, so will also apply to linux-next).

Is this fix needed in 2.6.26.x? 2.6.25.x?

Yes, it's needed in 2.6.26.x and 2.6.25, but I doubt it is critical :
only needed for systemtap users which use markers and rebuild their kernel after having changed a marker format string.

Quoted 7 lines
Quoted 4 lines

From: Roland McGrath <roland@redhat dot com>
From: Wenji Huang <[email protected]>
CC: [email protected]

whoa, what's all that about?

Quoted 9 lines

Patches can only have one Author: in git, and we indicate that in emails by putting a From: line right at the top of the changelog. (Additional credits can and should be mentioned in the changelog text of course). When that From: line is missing we take the authorship info from the email headers.

So according to the above, this patch has three authors. Geeze, even I can type faster than that!

hehehe :)

Quoted 2 lines

Who wrote it?

I took the diffs Roland and Wenji sent in separate emails and created this patch. They are the ones who actually identified the solution. So I guess I should be the author, with credits to them, if they are ok with that.

Mathieu

Quoted 32 lines
Quoted 31 lines
---
 scripts/Makefile.modpost |    1 +
 scripts/mod/modpost.c    |    3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)

Index: linux-2.6-lttng/scripts/Makefile.modpost
===================================================================

scripts/Makefile.modpost +1 -0
@@ -101,6 +101,7 @@ quiet_cmd_kernel-mod = MODPOST $@
       cmd_kernel-mod = $(modpost) $@
 
 vmlinux.o: FORCE
+	@rm -fr $(kernelmarkersfile)
 	$(call cmd,kernel-mod)
 
 # Declare generated files as targets for modpost

Index: linux-2.6-lttng/scripts/mod/modpost.c
===================================================================

scripts/mod/modpost.c +2 -1
@@ -1992,7 +1992,8 @@ static void read_markers(const char *fna
 			mod->skip = 1;
 		}
 
-		add_marker(mod, marker, fmt);
+		if (!mod->skip)
+			add_marker(mod, marker, fmt);
 	}
 	return;
 fail:
Signature · 2 lines
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68
#9 lore raw

Re: [PATCH] Fix markers duplicate modpost entry (update)

In reply to Andrew Morton

When a kernel was rebuilt, the previous Module.markers was not cleared. It caused markers with different format strings to appear as duplicates when a markers was changed. This problem is present since scripts/mod/modpost.c started to generate Module.markers,
commit b2e3e658b344c6bcfb8fb694100ab2f2b5b2edb0

It therefore applies to 2.6.25, 2.6.26 and linux-next.

I merely merged the patches from Roland, Wenji and Takashi here.

Credits to
Roland McGrath <[email protected]>
Wenji Huang <[email protected]>
and
Takashi Nishiie <[email protected]>

for providing the individual fixes.

- Changelog :
- Integrated Takashi's Makefile modification to clear Module.markers upon make clean.

CC Roland McGrath <[email protected]>
CC Wenji Huang <[email protected]>
CC Takashi Nishiie <[email protected]>
---
 Makefile                 |    3 ++-
 scripts/Makefile.modpost |    1 +
 scripts/mod/modpost.c    |    3 ++-
 3 files changed, 5 insertions(+), 2 deletions(-)
scripts/Makefile.modpost +1 -0
Index: linux-2.6-lttng/scripts/Makefile.modpost
===================================================================
@@ -101,6 +101,7 @@ quiet_cmd_kernel-mod = MODPOST $@
       cmd_kernel-mod = $(modpost) $@
 
 vmlinux.o: FORCE
+	@rm -fr $(kernelmarkersfile)
 	$(call cmd,kernel-mod)
 
 # Declare generated files as targets for modpost
scripts/mod/modpost.c +2 -1
Index: linux-2.6-lttng/scripts/mod/modpost.c
===================================================================
@@ -1992,7 +1992,8 @@ static void read_markers(const char *fna
 			mod->skip = 1;
 		}
 
-		add_marker(mod, marker, fmt);
+		if (!mod->skip)
+			add_marker(mod, marker, fmt);
 	}
 	return;
 fail:
Makefile +2 -1
Index: linux-2.6-lttng/Makefile
===================================================================
@@ -1133,7 +1133,8 @@ clean: archclean $(clean-dirs)
 	@find . $(RCS_FIND_IGNORE) \
 		\( -name '*.[oas]' -o -name '*.ko' -o -name '.*.cmd' \
 		-o -name '.*.d' -o -name '.*.tmp' -o -name '*.mod.c' \
-		-o -name '*.symtypes' -o -name 'modules.order' \) \
+		-o -name '*.symtypes' -o -name 'modules.order' \
+		-o -name 'Module.markers' \) \
 		-type f -print | xargs rm -f
 
 # mrproper - Delete all generated files, including .config
Signature · 2 lines
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68
#10 lore raw

Re: [PATCH] Fix markers duplicate modpost entry (update)

Quoted 1 line · On Thu, 17 Jul 2008 10:52:17 -0400 Mathieu Desnoyers…

Subject: Re: [PATCH] Fix markers duplicate modpost entry (update)

Please prepare the patch titles along the lines of

[patch] subsystem-identifer: what i did to it

so this one would have been

[patch] markers: fix markers duplicate modpost entry

It's for consistency, and for ease of reading commits mailing lists and gitk titles and git-shortlog output, etc.

Thanks.

#11 lore raw

Re: [PATCH] Fix markers duplicate modpost entry (update)

Quoted 9 lines · On Thu, 17 Jul 2008 15:28:37 -0700 Andrew Morton wrote:
Quoted 1 line · On Thu, 17 Jul 2008 10:52:17 -0400 Mathieu Desnoyers…

Subject: Re: [PATCH] Fix markers duplicate modpost entry (update)

Please prepare the patch titles along the lines of

[patch] subsystem-identifer: what i did to it

Quoted 7 lines

so this one would have been

[patch] markers: fix markers duplicate modpost entry

It's for consistency, and for ease of reading commits mailing lists and gitk titles and git-shortlog output, etc.

Agreed, although I would have used

[PATCH] modpost: fix duplicate markers entries

---
~Randy
Linux Plumbers Conference, 17-19 September 2008, Portland, Oregon USA
http://linuxplumbersconf.org/