RE: [patch] x86_64 specific function return probes

["Lynch, Rusty" <rusty.lynch@intel.com>]

From: Andrew Morton [mailto:akpm@osdl.org]
>"Lynch, Rusty" <rusty.lynch@intel.com> wrote:
>>
>>
>> From: Andi Kleen [mailto:ak@suse.de]
>> >On Thu, Jun 02, 2005 at 09:09:09AM -0700, Rusty Lynch wrote:
>> >> The following patch adds the x86_64 architecture specific
>> implementation
>> >> for function return probes to the 2.6.12-rc5-mm2 kernel.
>> >
>> >This is not a sufficient description for a patch. Can you describe
>> >how it actually works and what it does?
>> >
>>
>> Ok, let me write up a description and I'll repost.
>
>You did, but:
>
>> >> + * Called when we hit the probe point at kretprobe_trampoline
>> >> + */
>> >> +int trampoline_probe_handler(struct kprobe *p, struct pt_regs
*regs)
>> >> +{
>> >> +	struct task_struct *tsk;
>> >> +	struct kretprobe_instance *ri;
>> >> +	struct hlist_head *head;
>> >> +	struct hlist_node *node;
>> >> +	unsigned long *sara = (unsigned long *)regs->rsp - 1;
>> >> +
>> >> +	tsk = arch_get_kprobe_task(sara);
>> >
>> >I dont think you handle the case of the exception happening on
>> >a exception or interrupt stack. This is broken.
>
>What about this problem?

I miss understood what Andi was pointing out, and thought the additional
description would clear things up.  Now I get what he is pointing out
and will reply to Andi's second email.

    --rusty
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/