restore ARMv6 OProfile support
Sorry, hit the wrong key. 'T' and 'Y' are too close together. (mutt)
----- Forwarded message from Russell King <[email protected]> -----
Date: Fri, 28 Dec 2007 19:57:24 +0000
From: Russell King <[email protected]>
To: Adrian Bunk <[email protected]>
Cc: Mathieu Desnoyers <[email protected]>,
Randy Dunlap <[email protected]>, [email protected], [email protected], [email protected]
Subject: Re: [2.6.24 patch] restore ARMv6 OProfile support
Quoted 4 lines · Adrian Bunk
This patch restores the ARMv6 OProfile support that was killed by commit
09cadedbdc01f1a4bea1f427d4fb4642eaa19da9.Signed-off-by Adrian Bunk <[email protected]>
Given where we are in the release cycle, I think this "cleanup" patch should be reverted. Once the issues with it are resolved (why was it never even CC'd to the arch maintainers for review?) then it can go into mainline.
Note that it also looks like (from the commitdiff) that the above mentioned commit also removes:
CONFIG_HARDWARE_PM (blackfin)CONFIG_OPROFILE_CELL (powerpc)
So they're probably subtly broken as well.
Linus, what do you think? Should 09cadedbdc01f1a4bea1f427d4fb4642eaa19da9 be reverted?
Quoted 29 lines
---
kernel/Kconfig.instrumentation | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
7fc221ef169610b5eac98e2ddd641811c0d53e4akernel/Kconfig.instrumentation +15
-0@@ -29,2 +29,17 @@ config OPROFILE +config OPROFILE_ARMV6 + bool + depends on OPROFILE && ARM && CPU_V6 && !SMP + default y + select OPROFILE_ARM11_CORE + +config OPROFILE_MPCORE + bool + depends on OPROFILE && ARM && CPU_V6 && SMP + default y + select OPROFILE_ARM11_CORE + +config OPROFILE_ARM11_CORE + bool + config KPROBES
Signature · 10 lines
Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: ----- End forwarded message ----- -- Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of:
I never got a response on my message, but I have just receieved:
| Date: Tue, 15 Jan 2008 11:05:00 +0100 | From: Joerg Wagner <[email protected]> | To: ARM Linux Mailing List <[email protected]> | Subject: 2.6.24-rc7 : oprofile on MPCore broken | | Hello, | | just tried to use oprofile on 2.6.24-rc7. | It does not detect the right processor | (/dev/oprofile/cpu_type contains "timer"). | | As I don't know exactly, how the string | "arm/mpcore" from arch/arm/oprofile/op_model_mpcore.c | gets feeded into that file, maybe someone else can help ?
So people are hitting the resulting mess created by 09cadedbdc01f1a4bea1f427d4fb4642eaa19da9. Can we please fix this regression one way or another please?
I don't particularly like stuffing the options into some random place in the architectures Kconfig file when they should stay along side the instrumentation configuration entries.
Quoted 70 lines · Russell King
Sorry, hit the wrong key. 'T' and 'Y' are too close together. (mutt)
----- Forwarded message from Russell King <[email protected]> -----
Date: Fri, 28 Dec 2007 19:57:24 +0000
From: Russell King <[email protected]>
To: Adrian Bunk <[email protected]>
Cc: Mathieu Desnoyers <[email protected]>,
Randy Dunlap <[email protected]>, [email protected], [email protected], [email protected]
Subject: Re: [2.6.24 patch] restore ARMv6 OProfile supportQuoted 4 lines · Adrian Bunk
This patch restores the ARMv6 OProfile support that was killed by commit
09cadedbdc01f1a4bea1f427d4fb4642eaa19da9.Signed-off-by Adrian Bunk <[email protected]>Given where we are in the release cycle, I think this "cleanup" patch should be reverted. Once the issues with it are resolved (why was it never even CC'd to the arch maintainers for review?) then it can go into mainline.
Note that it also looks like (from the commitdiff) that the above mentioned commit also removes:
CONFIG_HARDWARE_PM(blackfin)CONFIG_OPROFILE_CELL(powerpc)So they're probably subtly broken as well.
Linus, what do you think? Should
09cadedbdc01f1a4bea1f427d4fb4642eaa19da9be reverted?Quoted 29 lines
---
kernel/Kconfig.instrumentation | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
7fc221ef169610b5eac98e2ddd641811c0d53e4akernel/Kconfig.instrumentation +15
-0@@ -29,2 +29,17 @@ config OPROFILE +config OPROFILE_ARMV6 + bool + depends on OPROFILE && ARM && CPU_V6 && !SMP + default y + select OPROFILE_ARM11_CORE + +config OPROFILE_MPCORE + bool + depends on OPROFILE && ARM && CPU_V6 && SMP + default y + select OPROFILE_ARM11_CORE + +config OPROFILE_ARM11_CORE + bool + config KPROBESSignature · 5 lines
Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: ----- End forwarded message -----
Signature · 3 lines
Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of:
Quoted 17 lines · Russell King
I never got a response on my message, but I have just receieved:
| Date: Tue, 15 Jan 2008 11:05:00 +0100 | From: Joerg Wagner <[email protected]> | To: ARM Linux Mailing List <[email protected]> | Subject: 2.6.24-rc7 : oprofile on MPCore broken | | Hello, | | just tried to use oprofile on 2.6.24-rc7. | It does not detect the right processor | (/dev/oprofile/cpu_type contains "timer"). | | As I don't know exactly, how the string | "arm/mpcore" from arch/arm/oprofile/op_model_mpcore.c | gets feeded into that file, maybe someone else can help ?
Quoted 6 lines
So people are hitting the resulting mess created by
09cadedbdc01f1a4bea1f427d4fb4642eaa19da9. Can we please fix this regression one way or another please?I don't particularly like stuffing the options into some random place in the architectures
Kconfigfile when they should stay along side the instrumentation configuration entries.
Below is the patch I already sent on 28 Dec 2007 that stuffs it into Kconfig.instrumentation.
Technically it shouldn't make any difference whether this patch or Mathieu's patch that stuffs it into arch/arm/Kconfig gets applied, but one of them should be applied for 2.6.24 (plus either mine or Mathieu's fix for the blackfin HARDWARE_PM support broken by the same commit).
I found the bugs in Mathieu's commit 09cadedbdc01f1a4bea1f427d4fb4642eaa19da9, I wrote patches that restore the status quo, Cc'ed all people even remotely related to this issue, and I opened the Bugzilla bugs required for getting them on the regression lists. Mathieu wants the regressions he introduced fixed different from what my patches did and that's not a problem for me (his patches are also OK).
What went wrong that his regression fixes did not land in Linus' tree?
cu
Adrian
<-- snip -->
This patch restores the ARMv6 OProfile support that was killed by commit 09cadedbdc01f1a4bea1f427d4fb4642eaa19da9.
---
kernel/Kconfig.instrumentation | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
7fc221ef169610b5eac98e2ddd641811c0d53e4a
kernel/Kconfig.instrumentation +15 -0
@@ -29,2 +29,17 @@ config OPROFILE +config OPROFILE_ARMV6 + bool + depends on OPROFILE && ARM && CPU_V6 && !SMP + default y + select OPROFILE_ARM11_CORE + +config OPROFILE_MPCORE + bool + depends on OPROFILE && ARM && CPU_V6 && SMP + default y + select OPROFILE_ARM11_CORE + +config OPROFILE_ARM11_CORE + bool + config KPROBES
Quoted 33 lines · * Adrian Bunk ([email protected]) wrote:
Quoted 17 lines · Russell King
I never got a response on my message, but I have just receieved:
| Date: Tue, 15 Jan 2008 11:05:00 +0100 | From: Joerg Wagner <[email protected]> | To: ARM Linux Mailing List <[email protected]> | Subject: 2.6.24-rc7 : oprofile on MPCore broken | | Hello, | | just tried to use oprofile on 2.6.24-rc7. | It does not detect the right processor | (/dev/oprofile/cpu_type contains "timer"). | | As I don't know exactly, how the string | "arm/mpcore" from arch/arm/oprofile/op_model_mpcore.c | gets feeded into that file, maybe someone else can help ?Quoted 6 lines
So people are hitting the resulting mess created by
09cadedbdc01f1a4bea1f427d4fb4642eaa19da9. Can we please fix this regression one way or another please?I don't particularly like stuffing the options into some random place in the architectures
Kconfigfile when they should stay along side the instrumentation configuration entries.Below is the patch I already sent on 28 Dec 2007 that stuffs it into
Kconfig.instrumentation.Technically it shouldn't make any difference whether this patch or Mathieu's patch that stuffs it into
arch/arm/Kconfiggets applied, but one of them should be applied for 2.6.24 (plus either mine or Mathieu's fix for the blackfinHARDWARE_PMsupport broken by the same commit).
Quoted 9 lines
I found the bugs in Mathieu's commit
09cadedbdc01f1a4bea1f427d4fb4642eaa19da9, I wrote patches that restore the status quo, Cc'ed all people even remotely related to this issue, and I opened the Bugzilla bugs required for getting them on the regression lists. Mathieu wants the regressions he introduced fixed different from what my patches did and that's not a problem for me (his patches are also OK).What went wrong that his regression fixes did not land in Linus' tree?
Yes Adrian, and I thank you very much for all this work. I only suggested the alternative patches so it would remove menu options nobody really needs (historical special-cases) and make it easier to apply the following set of patches already in -mm.
One way or another, either your or my arm and blackfin patches should get into 2.6.24 final.
Mathieu
Quoted 40 lines
cu
Adrian<-- snip -->
This patch restores the ARMv6 OProfile support that was killed by commit
09cadedbdc01f1a4bea1f427d4fb4642eaa19da9.Signed-off-by Adrian Bunk <[email protected]>---
kernel/Kconfig.instrumentation | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
7fc221ef169610b5eac98e2ddd641811c0d53e4akernel/Kconfig.instrumentation +15
-0@@ -29,2 +29,17 @@ config OPROFILE +config OPROFILE_ARMV6 + bool + depends on OPROFILE && ARM && CPU_V6 && !SMP + default y + select OPROFILE_ARM11_CORE + +config OPROFILE_MPCORE + bool + depends on OPROFILE && ARM && CPU_V6 && SMP + default y + select OPROFILE_ARM11_CORE + +config OPROFILE_ARM11_CORE + bool + config KPROBES
Signature · 3 lines
Mathieu Desnoyers Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
Quoted 4 lines · Russell King
I don't particularly like stuffing the options into some random place in the architectures
Kconfigfile when they should stay along side the instrumentation configuration entries.
Well, I have to say that I don't particularly like obviously architecture-specific stuff in an obviously non-architecture file..
I'd almost prefer to revert the thing that caused the problem, because with Adrian's patch, I think the end result may *work*, but it's uglier than what we started out with.
However, I think the *cleanest* solution right now may be something like the appended. Totally untested, of course. It basically just copies the generic kernel/Kconfig.instrumentation file into the arm directory, makes arm use its own instead of the generic one, and removes the dependencies on ARM in there (including all of the KPROBES entry that apparently isn't an issue on ARM anyway). It then adds back the ARM-specific ones.
This follows the sacred rules of good code:
- generic code is either generic or not. If it's not generic, don't claim it is.
- don't *force* people to use generic code if it doesn't suit them. Make it available for the cases it makes sense for, but don't shoe-horn it into cases where it doesn't work well.
So it allows the sharing of the common case and *many* architectures end up using the generic Kconfig file, but hey, if it doesn't make sense for ARM, it doesn't make sense for ARM. It's that simple.
But as mentioned, it's totally untested and I don't have (or really want to have) a cross-compiling environment. And I don't care *that* much. I just want something we can all live with.
So does something like this work for people?
Linus
--- arch/arm/Kconfig | 2 +- arch/arm/Kconfig.instrumentation | 52 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 53 insertions(+), 1 deletions(-)
arch/arm/Kconfig +1 -1
@@ -1076,7 +1076,7 @@ endmenu source "fs/Kconfig" -source "kernel/Kconfig.instrumentation" +source "arch/arm/Kconfig.instrumentation" source "arch/arm/Kconfig.debug"
arch/arm/Kconfig.instrumentation new file +52 -0
52 lines not on this page · raw message
Ingo Molnar
Acked-byQuoted 9 lines · Linus Torvalds
But as mentioned, it's totally untested and I don't have (or really want to have) a cross-compiling environment. And I don't care *that* much. I just want something we can all live with.
So does something like this work for people?
arch/arm/Kconfig | 2 +- arch/arm/Kconfig.instrumentation | 52 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 53 insertions(+), 1 deletions(-)
i like this approach better, not the least because it affects only one architecture so late in the .24-rc cycle. Instrumentation bugs tend to be found with a few weeks/months of delays. (because historically instrumentation has been typically used by folks who cannot change their kernel easily to debug it, i.e. by extension they dont run bleeding edge kernels either.)
Ingo
Quoted 1 line · Linus Torvalds
So does something like this work for people?
Just tested it and the right Kconfig symbols get set again - thanks.
Signature · 3 lines
Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of:
Quoted 35 lines · * Linus Torvalds ([email protected]) wrote:
Quoted 4 lines · Russell King
I don't particularly like stuffing the options into some random place in the architectures
Kconfigfile when they should stay along side the instrumentation configuration entries.Well, I have to say that I don't particularly like obviously architecture-specific stuff in an obviously non-architecture file..
I'd almost prefer to revert the thing that caused the problem, because with Adrian's patch, I think the end result may *work*, but it's uglier than what we started out with.
However, I think the *cleanest* solution right now may be something like the appended. Totally untested, of course. It basically just copies the generic
kernel/Kconfig.instrumentationfile into the arm directory, makes arm use its own instead of the generic one, and removes the dependencies on ARM in there (including all of the KPROBES entry that apparently isn't an issue on ARM anyway). It then adds back the ARM-specific ones.This follows the sacred rules of good code:
- generic code is either generic or not. If it's not generic, don't claim it is.
- don't *force* people to use generic code if it doesn't suit them. Make it available for the cases it makes sense for, but don't shoe-horn it into cases where it doesn't work well.
So it allows the sharing of the common case and *many* architectures end up using the generic
Kconfigfile, but hey, if it doesn't make sense for ARM, it doesn't make sense for ARM. It's that simple.
Quoted 6 lines
But as mentioned, it's totally untested and I don't have (or really want to have) a cross-compiling environment. And I don't care *that* much. I just want something we can all live with.
So does something like this work for people?
Hi,
Well, it goes along the lines of the patch I suggested as a reply to Adrian, with these differences :
- I still source the kernel/Kconfig.instrumentation file.
- I put back the missing OPROFILE options directly in arch/arm/Kconfig
Then end result is the same as your patch, but without the code duplication.
Here is the patch :
Fix ARMv6 oprofile support
This patch restores the ARMv6 OProfile support that was killed by commit 09cadedbdc01f1a4bea1f427d4fb4642eaa19da9.
It puts the config options in arch/arm/Kconfig.
Thanks to Adrian Bunk for finding this bug and providing an initial patch.
Changelog :
Use def_bool.
--- arch/arm/Kconfig | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)
arch/arm/Kconfig +17 -0
@@ -130,6 +130,23 @@ config FIQ config ARCH_MTD_XIP bool +if OPROFILE + +config OPROFILE_ARMV6 + def_bool y + depends on CPU_V6 && !SMP + select OPROFILE_ARM11_CORE + +config OPROFILE_MPCORE + def_bool y + depends on CPU_V6 && SMP + select OPROFILE_ARM11_CORE + +config OPROFILE_ARM11_CORE + bool + +endif + config VECTORS_BASE hex default 0xffff0000 if MMU || CPU_HIGH_VECTOR
-
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
Quoted 9 lines · Mathieu Desnoyers
Well, it goes along the lines of the patch I suggested as a reply to Adrian, with these differences :
- I still source the
kernel/Kconfig.instrumentationfile.
- I put back the missing OPROFILE options directly inarch/arm/KconfigThen end result is the same as your patch, but without the code duplication.
No it's not.
Now the config variables may all be there, but the UI for the *menu* system is broken (ie all the ARM profiling config options are now outside the profiling menu).
Is that menu really needed? I dunno. But since it exists, it should be correct.
Linus
Quoted 15 lines · * Linus Torvalds ([email protected]) wrote:
Quoted 9 lines · Mathieu Desnoyers
Well, it goes along the lines of the patch I suggested as a reply to Adrian, with these differences :
- I still source the
kernel/Kconfig.instrumentationfile.
- I put back the missing OPROFILE options directly inarch/arm/KconfigThen end result is the same as your patch, but without the code duplication.
No it's not.
Quoted 8 lines
Now the config variables may all be there, but the UI for the *menu* system is broken (ie all the ARM profiling config options are now outside the profiling menu).
Is that menu really needed? I dunno. But since it exists, it should be correct.
Linus
There is an "instrumentation menu removal" patchset I've submitted to Andrew for the next release cycle that moves the instrumentation menu content into General setup (I did this following your advice).
Furthermore, on ARM, the OPROFILE_ARMV6, OPROFILE_MPCORE and OPROFILE_ARM11_CORE are all "bool , default y" (equivalent to the preferred def_bool y). Unless I am grossly mistaken, this is not supposed to show up in the menus; it's just selected when the dependencies are met.
Mathieu
Signature · 3 lines
Mathieu Desnoyers Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
Quoted 29 lines · Mathieu Desnoyers
Quoted 15 lines · * Linus Torvalds ([email protected]) wrote:
Quoted 9 lines · Mathieu Desnoyers
Well, it goes along the lines of the patch I suggested as a reply to Adrian, with these differences :
- I still source the
kernel/Kconfig.instrumentationfile.
- I put back the missing OPROFILE options directly inarch/arm/KconfigThen end result is the same as your patch, but without the code duplication.
No it's not.
Quoted 8 lines
Now the config variables may all be there, but the UI for the *menu* system is broken (ie all the ARM profiling config options are now outside the profiling menu).
Is that menu really needed? I dunno. But since it exists, it should be correct.
Linus
There is an "instrumentation menu removal" patchset I've submitted to Andrew for the next release cycle that moves the instrumentation menu content into General setup (I did this following your advice).
Quoted 5 lines
Furthermore, on ARM, the
OPROFILE_ARMV6,OPROFILE_MPCOREandOPROFILE_ARM11_COREare all "bool , default y" (equivalent to the preferreddef_booly). Unless I am grossly mistaken, this is not supposed to show up in the menus; it's just selected when the dependencies are met.
This was the patch:
+if OPROFILE
+
+config OPROFILE_ARMV6
+ def_bool y
+ depends on CPU_V6 && !SMP
+ select OPROFILE_ARM11_CORE
+
+config OPROFILE_MPCORE
+ def_bool y
+ depends on CPU_V6 && SMP
+ select OPROFILE_ARM11_CORE
+
+config OPROFILE_ARM11_CORE
+ bool
+
+endif
And none of these has a prompt defined so they do not show up as a menu.
It is very easy to test if you have the patch applied. No cross toolchin is needed - just do:
make ARCH=arm menuconfig
Sam
Quoted 54 lines · * Sam Ravnborg ([email protected]) wrote:
Quoted 29 lines · Mathieu Desnoyers
Quoted 15 lines · * Linus Torvalds ([email protected]) wrote:
Quoted 9 lines · Mathieu Desnoyers
Well, it goes along the lines of the patch I suggested as a reply to Adrian, with these differences :
- I still source the
kernel/Kconfig.instrumentationfile.
- I put back the missing OPROFILE options directly inarch/arm/KconfigThen end result is the same as your patch, but without the code duplication.
No it's not.
Quoted 8 lines
Now the config variables may all be there, but the UI for the *menu* system is broken (ie all the ARM profiling config options are now outside the profiling menu).
Is that menu really needed? I dunno. But since it exists, it should be correct.
Linus
There is an "instrumentation menu removal" patchset I've submitted to Andrew for the next release cycle that moves the instrumentation menu content into General setup (I did this following your advice).
Quoted 5 lines
Furthermore, on ARM, the
OPROFILE_ARMV6,OPROFILE_MPCOREandOPROFILE_ARM11_COREare all "bool , default y" (equivalent to the preferreddef_booly). Unless I am grossly mistaken, this is not supposed to show up in the menus; it's just selected when the dependencies are met.This was the patch:
arch/arm/Kconfig +16
-0+if OPROFILE + +config OPROFILE_ARMV6 + def_bool y + depends on CPU_V6 && !SMP + select OPROFILE_ARM11_CORE + +config OPROFILE_MPCORE + def_bool y + depends on CPU_V6 && SMP + select OPROFILE_ARM11_CORE + +config OPROFILE_ARM11_CORE + bool + +endif
Quoted 7 lines
And none of these has a prompt defined so they do not show up as a menu.
It is very easy to test if you have the patch applied. No cross toolchin is needed - just do:
make ARCH=arm menuconfigSam
Just tested with and without oprofile/MPCORE/CPU_V6 combinations and I confirm that :
- no menu is showing up (as expected)
- the OPROFILE_* config options are selected (or not) as expected
Mathieu
Signature · 3 lines
Mathieu Desnoyers Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
Quoted 4 lines · Mathieu Desnoyers
Furthermore, on ARM, the
OPROFILE_ARMV6,OPROFILE_MPCOREandOPROFILE_ARM11_COREare all "bool , default y" (equivalent to the preferreddef_booly).
Ahh, I didn't notice that. If so, yes, it doesn't matter at all whether it's inside the menu or not, of course.
Linus
Quoted 35 lines · Linus Torvalds
Quoted 4 lines · Russell King
I don't particularly like stuffing the options into some random place in the architectures
Kconfigfile when they should stay along side the instrumentation configuration entries.Well, I have to say that I don't particularly like obviously architecture-specific stuff in an obviously non-architecture file..
I'd almost prefer to revert the thing that caused the problem, because with Adrian's patch, I think the end result may *work*, but it's uglier than what we started out with.
However, I think the *cleanest* solution right now may be something like the appended. Totally untested, of course. It basically just copies the generic
kernel/Kconfig.instrumentationfile into the arm directory, makes arm use its own instead of the generic one, and removes the dependencies on ARM in there (including all of the KPROBES entry that apparently isn't an issue on ARM anyway). It then adds back the ARM-specific ones.This follows the sacred rules of good code:
- generic code is either generic or not. If it's not generic, don't claim it is.
- don't *force* people to use generic code if it doesn't suit them. Make it available for the cases it makes sense for, but don't shoe-horn it into cases where it doesn't work well.
So it allows the sharing of the common case and *many* architectures end up using the generic
Kconfigfile, but hey, if it doesn't make sense for ARM, it doesn't make sense for ARM. It's that simple.
Quoted 5 lines
But as mentioned, it's totally untested and I don't have (or really want to have) a cross-compiling environment. And I don't care *that* much. I just want something we can all live with.
So does something like this work for people?
BTW, your patch may fix ARM, but the original commit broke blackfin as well - it removed the Kconfig entry for HARDWARE_PM, which is still used:
$ grep HARDWARE_PM arch/blackfin/ -r arch/blackfin/mach-common/interrupt.S:#ifdef CONFIG_HARDWARE_PM arch/blackfin/mach-common/interrupt.S:#ifdef CONFIG_HARDWARE_PM arch/blackfin/mach-common/irqpanic.c:#ifdef CONFIG_HARDWARE_PM arch/blackfin/oprofile/Makefile:oprofile-$(CONFIG_HARDWARE_PM) += op_model_bf533.o arch/blackfin/oprofile/common.c:#ifdef CONFIG_HARDWARE_PM
So blackfin also needs fixing. I don't know if blackfin people are aware of this.
Signature · 3 lines
Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of:
Quoted 3 lines · Russell King
...
So blackfin also needs fixing. I don't know if blackfin people are aware of this.
They are:
http://lkml.org/lkml/2007/12/28/100
Quoted 1 line
Russell King
cu
Adrian
Signature · 4 lines
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed