mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: zhang warden <zhangwarden@gmail.com>
To: Miroslav Benes <mbenes@suse.cz>
Cc: Josh Poimboeuf <jpoimboe@kernel.org>,
	Jiri Kosina <jikos@kernel.org>, Petr Mladek <pmladek@suse.com>,
	Joe Lawrence <joe.lawrence@redhat.com>,
	live-patching@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v4 1/2] Introduce klp_ops into klp_func structure
Date: Fri, 13 Sep 2024 17:46:18 +0800	[thread overview]
Message-ID: <D2C635CE-4530-492C-B9CF-CFF78FE116A1@gmail.com> (raw)
In-Reply-To: <7F83CD32-3965-4F15-B4FA-44503EF6EA9D@gmail.com>

Hi, Miroslav & Petr

> On Sep 6, 2024, at 17:44, zhang warden <zhangwarden@gmail.com> wrote:
> 
>> 
>>>> 
>>>>> +       struct list_head func_stack;
>>>>> +       struct ftrace_ops fops;
>>>>> +};
>>>>> +
>>>>> 
>>>>> diff --git a/kernel/livepatch/core.c b/kernel/livepatch/core.c
>>>>> index 52426665eecc..e4572bf34316 100644
>>>>> --- a/kernel/livepatch/core.c
>>>>> +++ b/kernel/livepatch/core.c
>>>>> @@ -760,6 +760,8 @@ static int klp_init_func(struct klp_object *obj, struct klp_func *func)
>>>>> if (!func->old_name)
>>>>> return -EINVAL;
>>>>> 
>>>>> + func->ops = NULL;
>>>>> +
>>>> 
>>>> Any reason why it is not added a couple of lines later alongside the rest 
>>>> of the initialization?
>>> 
>>> Do you mean I should add couple of lines after 'return -EINVAL' ?
>> 
>> No, I am asking if there is a reason why you added 'func->ops = NULL;' 
>> here and not right after the rest of func initializations
>> 
>>       INIT_LIST_HEAD(&func->stack_node);
>>       func->patched = false;
>>       func->transition = false;
>> 
> 

I think I found a bug in my patch.

I move struct klp_ops to klp_func. But every time, klp_func
will init klp_func->ops to NULL. Which will make the test branch
 `if(! ops)` always true in function klp_patch_func.

An alternative solution should be something like
 (as Peter suggested before) 

https://lore.kernel.org/live-patching/20240805064656.40017-1-zhangyongde.zyd@alibaba-inc.com/T/#t


 static struct klp_ops *klp_find_ops(void *old_func)
 {
       struct klp_patch *patch;
       struct klp_object *obj;
       struct klp_func *func;
       struct klp_ops *ops;
 
        klp_for_each_patch(patch)
                klp_for_each_object(patch, obj)
                        klp_for_each_func(obj, func)
                               if(func->old_func == old_func)
                                        return func->ops;
        return NULL;
 }

and  klp_ops should be initialize in klp_init_func:

        func->patched = false;
        func->transition = false;
+       func->ops = klp_find_ops(func->old_func);

Here, func->ops should not init as NULL, it should be initialize with 
the existed ops (if klp_find_ops returns NULL, this patch is the first
time to be patched).

Regards.
Wardenjohn.


  reply	other threads:[~2024-09-13  9:46 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-28  2:23 [PATCH v4 0/2] livepatch: Add using attribute to klp_func for using function Wardenjohn
2024-08-28  2:23 ` [PATCH v4 1/2] Introduce klp_ops into klp_func structure Wardenjohn
2024-09-05 10:10   ` Miroslav Benes
2024-09-05 14:33     ` zhang warden
2024-09-06  7:03       ` Miroslav Benes
2024-09-06  9:44         ` zhang warden
2024-09-13  9:46           ` zhang warden [this message]
2024-08-28  2:23 ` [PATCH v4 2/2] livepatch: Add using attribute to klp_func for using function show Wardenjohn
2024-09-04  1:54   ` zhang warden
2024-09-04  4:48   ` Josh Poimboeuf
2024-09-04  6:34     ` zhang warden
2024-09-04  7:14       ` Josh Poimboeuf
2024-09-04  7:30         ` zhang warden
2024-09-04 18:06           ` Josh Poimboeuf
2024-09-05 14:03             ` zhang warden
2024-09-05 16:30               ` Josh Poimboeuf
2024-09-05 10:23   ` Miroslav Benes
2024-09-05 14:17     ` zhang warden
2024-09-05 16:34     ` Josh Poimboeuf
2024-09-06  6:55       ` Miroslav Benes
2024-09-06  9:39       ` zhang warden
2024-09-06 16:39         ` Petr Mladek
2024-09-08  2:31           ` zhang warden
2024-09-06 16:13     ` Petr Mladek
2024-09-08  2:51       ` zhang warden
2024-09-10  8:01         ` Petr Mladek
2024-09-10  8:09           ` zhang warden

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=D2C635CE-4530-492C-B9CF-CFF78FE116A1@gmail.com \
    --to=zhangwarden@gmail.com \
    --cc=jikos@kernel.org \
    --cc=joe.lawrence@redhat.com \
    --cc=jpoimboe@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=live-patching@vger.kernel.org \
    --cc=mbenes@suse.cz \
    --cc=pmladek@suse.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

all inboxes | Powered by JetHome®