From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
To: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Sam Ravnborg <sam@ravnborg.org>,
akpm@linux-foundation.org, lkml <linux-kernel@vger.kernel.org>,
Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>,
davem@davemloft.net, hskinnemoen@atmel.com
Subject: Re: [PATCH 1/2] Kprobes: Indicate kretprobe support in arch/<arch>/Kconfig - updated
Date: Mon, 10 Dec 2007 10:10:01 -0500 [thread overview]
Message-ID: <20071210151001.GD2082@Krystal> (raw)
In-Reply-To: <20071210113249.GC8437@in.ibm.com>
* Ananth N Mavinakayanahalli (ananth@in.ibm.com) wrote:
> On Mon, Dec 10, 2007 at 11:13:07AM +0100, Sam Ravnborg wrote:
> > On Mon, Dec 10, 2007 at 03:22:22PM +0530, Ananth N Mavinakayanahalli wrote:
> > > From: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
> > >
> > > This patch adds CONFIG_HAVE_KRETPROBES to the arch/<arch>/Kconfig file
> > > for relevant architectures with kprobes support. This facilitates easy
> > > handling of in-kernel modules (like samples/kprobes/kretprobe_example.c)
> > > that depend on kretprobes being present in the kernel.
> > >
> > > This patch depends on Mathieu Desnoyers' "Instrumentation menu removal"
> > > patchset (http://marc.info/?l=linux-kernel&m=119496432229633&w=2)
> > >
> > > Updated to apply on 2.6.24-rc4-mm1.
> > >
> > > Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
> > > ---
> > > arch/avr32/Kconfig | 5 +++++
> > > arch/ia64/Kconfig | 5 +++++
> > > arch/powerpc/Kconfig | 5 +++++
> > > arch/s390/Kconfig | 5 +++++
> > > arch/sparc64/Kconfig | 5 +++++
> > > arch/x86/Kconfig | 5 +++++
> > > include/asm-ia64/kprobes.h | 1 -
> > > include/asm-powerpc/kprobes.h | 1 -
> > > include/asm-s390/kprobes.h | 1 -
> > > include/asm-x86/kprobes_32.h | 1 -
> > > include/asm-x86/kprobes_64.h | 1 -
> > > include/linux/kprobes.h | 6 +++---
> > > kernel/kprobes.c | 8 +++-----
> > > 13 files changed, 36 insertions(+), 13 deletions(-)
> > >
> > > Index: linux-2.6.24-rc4/arch/avr32/Kconfig
> > > ===================================================================
> > > --- linux-2.6.24-rc4.orig/arch/avr32/Kconfig
> > > +++ linux-2.6.24-rc4/arch/avr32/Kconfig
> > > @@ -66,6 +66,11 @@ config GENERIC_BUG
> > > def_bool y
> > > depends on BUG
> > >
> > > +config HAVE_KRETPROBES
> > > + bool
> > > + depends on HAVE_KPROBES
> > > + default n
> > > +
> >
> > The symbol HAVE_KRETPROBES should be defined in the common file - so
> > the only thing you have to do is a single line:
> > config AVR32
> > + select HAVE_KRETPROBES
> >
> >
> > Much more dense than defining the symbols once for each arch.
>
> Agreed. Updated patch below.
>
> From: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
>
> This patch adds CONFIG_HAVE_KRETPROBES to the arch/<arch>/Kconfig file
> for relevant architectures with kprobes support. This facilitates easy
> handling of in-kernel modules (like samples/kprobes/kretprobe_example.c)
> that depend on kretprobes being present in the kernel.
>
> This patch depends on Mathieu Desnoyers' "Instrumentation menu removal"
> patchset (http://marc.info/?l=linux-kernel&m=119496432229633&w=2)
>
> Updated to apply on 2.6.24-rc4-mm1. Thanks to Sam Ravnborg for helping
> make the patch more lean.
>
> Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
> ---
> arch/Kconfig | 4 ++++
> arch/ia64/Kconfig | 1 +
> arch/powerpc/Kconfig | 1 +
> arch/s390/Kconfig | 1 +
> arch/x86/Kconfig | 1 +
> include/asm-ia64/kprobes.h | 1 -
> include/asm-powerpc/kprobes.h | 1 -
> include/asm-x86/kprobes_32.h | 1 -
> include/asm-x86/kprobes_64.h | 1 -
> include/linux/kprobes.h | 6 +++---
> kernel/kprobes.c | 8 +++-----
> 11 files changed, 14 insertions(+), 12 deletions(-)
>
> Index: linux-2.6.24-rc4/arch/Kconfig
> ===================================================================
> --- linux-2.6.24-rc4.orig/arch/Kconfig
> +++ linux-2.6.24-rc4/arch/Kconfig
> @@ -29,3 +29,7 @@ config KPROBES
>
> config HAVE_KPROBES
> def_bool n
> +
> +config HAVE_KRETPROBES
> + def_bool n
> + depends on HAVE_KPROBES
> Index: linux-2.6.24-rc4/arch/ia64/Kconfig
> ===================================================================
> --- linux-2.6.24-rc4.orig/arch/ia64/Kconfig
> +++ linux-2.6.24-rc4/arch/ia64/Kconfig
> @@ -17,6 +17,7 @@ config IA64
> select ARCH_SUPPORTS_MSI
> select HAVE_OPROFILE
> select HAVE_KPROBES
> + select HAVE_KRETPROBES
> default y
> help
> The Itanium Processor Family is Intel's 64-bit successor to
> Index: linux-2.6.24-rc4/arch/powerpc/Kconfig
> ===================================================================
> --- linux-2.6.24-rc4.orig/arch/powerpc/Kconfig
> +++ linux-2.6.24-rc4/arch/powerpc/Kconfig
> @@ -81,6 +81,7 @@ config PPC
> default y
> select HAVE_OPROFILE
> select HAVE_KPROBES
> + select HAVE_KRETPROBES
>
> config EARLY_PRINTK
> bool
> Index: linux-2.6.24-rc4/arch/s390/Kconfig
> ===================================================================
> --- linux-2.6.24-rc4.orig/arch/s390/Kconfig
> +++ linux-2.6.24-rc4/arch/s390/Kconfig
> @@ -53,6 +53,7 @@ config S390
> def_bool y
> select HAVE_OPROFILE
> select HAVE_KPROBES
> + select HAVE_KRETPROBES
>
> source "init/Kconfig"
>
> Index: linux-2.6.24-rc4/arch/x86/Kconfig
> ===================================================================
> --- linux-2.6.24-rc4.orig/arch/x86/Kconfig
> +++ linux-2.6.24-rc4/arch/x86/Kconfig
> @@ -20,6 +20,7 @@ config X86
> def_bool y
> select HAVE_OPROFILE
> select HAVE_KPROBES
> + select HAVE_KRETPROBES
>
> config GENERIC_TIME
> def_bool y
> Index: linux-2.6.24-rc4/include/asm-ia64/kprobes.h
> ===================================================================
> --- linux-2.6.24-rc4.orig/include/asm-ia64/kprobes.h
> +++ linux-2.6.24-rc4/include/asm-ia64/kprobes.h
> @@ -82,7 +82,6 @@ struct kprobe_ctlblk {
> struct prev_kprobe prev_kprobe[ARCH_PREV_KPROBE_SZ];
> };
>
> -#define ARCH_SUPPORTS_KRETPROBES
> #define kretprobe_blacklist_size 0
>
> #define SLOT0_OPCODE_SHIFT (37)
> Index: linux-2.6.24-rc4/include/asm-powerpc/kprobes.h
> ===================================================================
> --- linux-2.6.24-rc4.orig/include/asm-powerpc/kprobes.h
> +++ linux-2.6.24-rc4/include/asm-powerpc/kprobes.h
> @@ -80,7 +80,6 @@ typedef unsigned int kprobe_opcode_t;
> #define is_trap(instr) (IS_TW(instr) || IS_TWI(instr))
> #endif
>
> -#define ARCH_SUPPORTS_KRETPROBES
> #define flush_insn_slot(p) do { } while (0)
> #define kretprobe_blacklist_size 0
>
> Index: linux-2.6.24-rc4/include/asm-x86/kprobes_32.h
> ===================================================================
> --- linux-2.6.24-rc4.orig/include/asm-x86/kprobes_32.h
> +++ linux-2.6.24-rc4/include/asm-x86/kprobes_32.h
> @@ -42,7 +42,6 @@ typedef u8 kprobe_opcode_t;
> ? (MAX_STACK_SIZE) \
> : (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR)))
>
> -#define ARCH_SUPPORTS_KRETPROBES
> #define flush_insn_slot(p) do { } while (0)
>
> extern const int kretprobe_blacklist_size;
> Index: linux-2.6.24-rc4/include/asm-x86/kprobes_64.h
> ===================================================================
> --- linux-2.6.24-rc4.orig/include/asm-x86/kprobes_64.h
> +++ linux-2.6.24-rc4/include/asm-x86/kprobes_64.h
> @@ -41,7 +41,6 @@ typedef u8 kprobe_opcode_t;
> ? (MAX_STACK_SIZE) \
> : (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR)))
>
> -#define ARCH_SUPPORTS_KRETPROBES
> extern const int kretprobe_blacklist_size;
>
> void kretprobe_trampoline(void);
> Index: linux-2.6.24-rc4/include/linux/kprobes.h
> ===================================================================
> --- linux-2.6.24-rc4.orig/include/linux/kprobes.h
> +++ linux-2.6.24-rc4/include/linux/kprobes.h
> @@ -125,11 +125,11 @@ struct jprobe {
> DECLARE_PER_CPU(struct kprobe *, current_kprobe);
> DECLARE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
>
> -#ifdef ARCH_SUPPORTS_KRETPROBES
> +#ifdef CONFIG_HAVE_KRETPROBES
Hi Ananth,
I just want to point out a detail: if someone sets CONFIG_KPROBES to n,
the CONFIG_HAVE_KPROBES is still y, and so is CONFIG_HAVE_KRETPROBES.
However, I doubt that you want to activate this code in this case ?
The code paths are OK because they are nested into CONFIG_KPROBES
ifdefs (or not built due to dependency on CONFIG_KPROBES in the
Makfile), but if one wants to use CONFIG_HAVE_KRETPROBE for something
else (Makefile), then it could become a problem.
Could we add a menu entry CONFIG_KRETPROBES that depends on
CONFIG_HAVE_KRETPROBES and CONFIG_KPROBES, and also remove the
CONFIG_HAVE_KPROBES dependency for the CONFIG_HAVE_KRETPROBE option ?
This way, we would have much more flexibility (like specifiying if we
want CONFIG_KRETPROBES to be default y or default n...)
Mathieu
> extern void arch_prepare_kretprobe(struct kretprobe_instance *ri,
> struct pt_regs *regs);
> extern int arch_trampoline_kprobe(struct kprobe *p);
> -#else /* ARCH_SUPPORTS_KRETPROBES */
> +#else /* CONFIG_HAVE_KRETPROBES */
> static inline void arch_prepare_kretprobe(struct kretprobe *rp,
> struct pt_regs *regs)
> {
> @@ -138,7 +138,7 @@ static inline int arch_trampoline_kprobe
> {
> return 0;
> }
> -#endif /* ARCH_SUPPORTS_KRETPROBES */
> +#endif /* CONFIG_HAVE_KRETPROBES */
> /*
> * Function-return probe -
> * Note:
> Index: linux-2.6.24-rc4/kernel/kprobes.c
> ===================================================================
> --- linux-2.6.24-rc4.orig/kernel/kprobes.c
> +++ linux-2.6.24-rc4/kernel/kprobes.c
> @@ -678,8 +678,7 @@ void __kprobes unregister_jprobe(struct
> unregister_kprobe(&jp->kp);
> }
>
> -#ifdef ARCH_SUPPORTS_KRETPROBES
> -
> +#ifdef CONFIG_HAVE_KRETPROBES
> /*
> * This kprobe pre_handler is registered with every kretprobe. When probe
> * hits it will set up the return probe.
> @@ -762,7 +761,7 @@ int __kprobes register_kretprobe(struct
> return ret;
> }
>
> -#else /* ARCH_SUPPORTS_KRETPROBES */
> +#else /* CONFIG_HAVE_KRETPROBES */
>
> int __kprobes register_kretprobe(struct kretprobe *rp)
> {
> @@ -774,8 +773,7 @@ static int __kprobes pre_handler_kretpro
> {
> return 0;
> }
> -
> -#endif /* ARCH_SUPPORTS_KRETPROBES */
> +#endif /* CONFIG_HAVE_KRETPROBES */
>
> void __kprobes unregister_kretprobe(struct kretprobe *rp)
> {
--
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
next prev parent reply other threads:[~2007-12-10 15:10 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-12-10 9:52 [PATCH 1/2] Kprobes: Indicate kretprobe support in arch/<arch>/Kconfig Ananth N Mavinakayanahalli
2007-12-10 9:53 ` [PATCH 2/2] Kprobes: Build kretprobe samples only if arch supports kretprobes Ananth N Mavinakayanahalli
2007-12-10 12:00 ` Sam Ravnborg
2007-12-10 13:42 ` [PATCH 2/2] Kprobes: Build kretprobe samples only if arch supports kretprobes - updated Ananth N Mavinakayanahalli
2007-12-10 10:13 ` [PATCH 1/2] Kprobes: Indicate kretprobe support in arch/<arch>/Kconfig Sam Ravnborg
2007-12-10 11:32 ` [PATCH 1/2] Kprobes: Indicate kretprobe support in arch/<arch>/Kconfig - updated Ananth N Mavinakayanahalli
2007-12-10 15:10 ` Mathieu Desnoyers [this message]
2007-12-11 6:09 ` Ananth N Mavinakayanahalli
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20071210151001.GD2082@Krystal \
--to=mathieu.desnoyers@polymtl.ca \
--cc=akpm@linux-foundation.org \
--cc=ananth@in.ibm.com \
--cc=anil.s.keshavamurthy@intel.com \
--cc=davem@davemloft.net \
--cc=hskinnemoen@atmel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=sam@ravnborg.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox