public inbox for linux-arch@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Christoph Hellwig <hch@infradead.org>,
	linux-kernel@vger.kernel.org,
	Prasanna S Panchamukhi <prasanna@in.ibm.com>,
	Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>,
	Jim Keniston <jkenisto@us.ibm.com>,
	linux-arch@vger.kernel.org
Subject: Re: [PATCH 2/3] Kprobes: Define retval helper
Date: Wed, 9 Aug 2006 10:45:16 +0100	[thread overview]
Message-ID: <20060809094516.GA17993@infradead.org> (raw)
In-Reply-To: <20060809094311.GA20050@in.ibm.com>

On Wed, Aug 09, 2006 at 03:13:11PM +0530, Ananth N Mavinakayanahalli wrote:
> > Good idea.  You should add parentheses around regs, otherwise the C
> > preprocessor might bite users.  Also the shouting name is quite ugly.
> > In fact it should probably go to asm/system.h or similar and not have
> > a kprobes name - it just extracts the return value from a struct pt_regs
> > after all.
> 
> Done! How does this look? I added it to asm/ptrace.h so it lives along
> with the instruction_pointer() definition.

Looks good, but it would be much better if we had it for every single
architecture.  I've cc'ed linux-arch so the arch maintainers can comments.

Even if we don't manage to get every architecture maintainer to help out
you should at least add sparc and s390 to have full coverage of
architectures with kprobes support

> Add the "get_retval" macro that just extracts the return value given the
> pt_regs. Useful in situations such as while using function-return
> probes.
> 
> Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
> 
> ---
>  include/asm-i386/ptrace.h    |    3 +++
>  include/asm-ia64/ptrace.h    |    3 +++
>  include/asm-powerpc/ptrace.h |    2 ++
>  include/asm-x86_64/ptrace.h  |    2 ++
>  4 files changed, 10 insertions(+)
> 
> Index: linux-2.6.18-rc3/include/asm-i386/ptrace.h
> ===================================================================
> --- linux-2.6.18-rc3.orig/include/asm-i386/ptrace.h
> +++ linux-2.6.18-rc3/include/asm-i386/ptrace.h
> @@ -79,7 +79,10 @@ static inline int user_mode_vm(struct pt
>  {
>  	return ((regs->xcs & 3) | (regs->eflags & VM_MASK)) != 0;
>  }
> +
>  #define instruction_pointer(regs) ((regs)->eip)
> +#define get_retval(regs) ((regs)->eax)
> +
>  #if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
>  extern unsigned long profile_pc(struct pt_regs *regs);
>  #else
> Index: linux-2.6.18-rc3/include/asm-ia64/ptrace.h
> ===================================================================
> --- linux-2.6.18-rc3.orig/include/asm-ia64/ptrace.h
> +++ linux-2.6.18-rc3/include/asm-ia64/ptrace.h
> @@ -237,6 +237,9 @@ struct switch_stack {
>   * the canonical representation by adding to instruction pointer.
>   */
>  # define instruction_pointer(regs) ((regs)->cr_iip + ia64_psr(regs)->ri)
> +
> +#define get_retval(regs) ((regs)->r8)
> +
>  /* Conserve space in histogram by encoding slot bits in address
>   * bits 2 and 3 rather than bits 0 and 1.
>   */
> Index: linux-2.6.18-rc3/include/asm-powerpc/ptrace.h
> ===================================================================
> --- linux-2.6.18-rc3.orig/include/asm-powerpc/ptrace.h
> +++ linux-2.6.18-rc3/include/asm-powerpc/ptrace.h
> @@ -73,6 +73,8 @@ struct pt_regs {
>  #ifndef __ASSEMBLY__
>  
>  #define instruction_pointer(regs) ((regs)->nip)
> +#define get_retval(regs) ((regs)->gpr[3])
> +
>  #ifdef CONFIG_SMP
>  extern unsigned long profile_pc(struct pt_regs *regs);
>  #else
> Index: linux-2.6.18-rc3/include/asm-x86_64/ptrace.h
> ===================================================================
> --- linux-2.6.18-rc3.orig/include/asm-x86_64/ptrace.h
> +++ linux-2.6.18-rc3/include/asm-x86_64/ptrace.h
> @@ -84,6 +84,8 @@ struct pt_regs {
>  #define user_mode(regs) (!!((regs)->cs & 3))
>  #define user_mode_vm(regs) user_mode(regs)
>  #define instruction_pointer(regs) ((regs)->rip)
> +#define get_retval(regs) ((regs)->rax)
> +
>  extern unsigned long profile_pc(struct pt_regs *regs);
>  void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
>  
---end quoted text---

       reply	other threads:[~2006-08-09  9:45 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20060807115537.GA15253@in.ibm.com>
     [not found] ` <20060807120024.GD15253@in.ibm.com>
     [not found]   ` <20060808162559.GB28647@infradead.org>
     [not found]     ` <20060809094311.GA20050@in.ibm.com>
2006-08-09  9:45       ` Christoph Hellwig [this message]
2006-08-09  9:55         ` [PATCH 2/3] Kprobes: Define retval helper Andi Kleen
2006-08-09 10:19           ` Ananth N Mavinakayanahalli
2006-08-09 10:28             ` Martin Schwidefsky
2006-08-09 13:16         ` David Howells
2006-08-09 15:20           ` 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=20060809094516.GA17993@infradead.org \
    --to=hch@infradead.org \
    --cc=ananth@in.ibm.com \
    --cc=anil.s.keshavamurthy@intel.com \
    --cc=jkenisto@us.ibm.com \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=prasanna@in.ibm.com \
    /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