This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
RE: Review patches of user space kprobe
- From: "Zhang, Yanmin" <yanmin dot zhang at intel dot com>
- To: <prasanna at in dot ibm dot com>
- Cc: <systemtap at sources dot redhat dot com>, "Keshavamurthy, Anil S" <anil dot s dot keshavamurthy at intel dot com>, "Mao, Bibo" <bibo dot mao at intel dot com>
- Date: Mon, 9 Jan 2006 09:48:22 +0800
- Subject: RE: Review patches of user space kprobe
>>-----Original Message-----
>>From: systemtap-owner@sourceware.org [mailto:systemtap-owner@sourceware.org] On Behalf Of Prasanna S Panchamukhi
>>Sent: 2006年1月6日 17:32
>>To: Zhang, Yanmin
>>Cc: systemtap@sources.redhat.com; Keshavamurthy, Anil S; Mao, Bibo
>>Subject: Re: Review patches of user space kprobe
>>
>>> >>> are only executed after the loop of hlist_for_each_entry. Is it correct?
>>> >>>
>>> >>
>>> >>that's correct. The page is mapped only once for the first match in the loop,
>>> >>then all the probes are inserted into that page in the hlist_for_each loop and
>>> >>then the page is unmapped only once after the end of the loop.
>>> So call lock_page for many times, and call unlock_page for one time?
>>>
>>no, lock_page and unlock_page is called only once, there is a small correction,
>>insert_probe_page() should be moved out of if condition so that all probes
>>are inserted on that page.
I'm still confused. Anyway, I will wait for your new codes. In addition, I still think the second parameter of map_uprobe_page is not good and could be deleted. The logic related to the second parameter is very simple and could be moved to the caller of map_uprobe_page.