From: Petr Mladek <pmladek@suse.com>
To: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: David Vernet <void@manifault.com>,
live-patching@vger.kernel.org, linux-kernel@vger.kernel.org,
jikos@kernel.org, mbenes@suse.cz, joe.lawrence@redhat.com,
corbet@lwn.net, songliubraving@fb.com,
gregkh@linuxfoundation.org
Subject: Re: [PATCH v2] livepatch: Fix leak on klp_init_patch_early failure path
Date: Mon, 20 Dec 2021 10:48:41 +0100 [thread overview]
Message-ID: <YcBReVgF2hFLpyjf@alley> (raw)
In-Reply-To: <20211217215042.76m5qn5e63ptgrjq@treble>
On Fri 2021-12-17 13:50:42, Josh Poimboeuf wrote:
> On Fri, Dec 17, 2021 at 02:51:42PM +0100, Petr Mladek wrote:
> > On Wed 2021-12-15 07:20:04, David Vernet wrote:
> > Josh's proposal adds pre-early-init() to allow calling
> > klp_free_patch_*() already when klp_init_*_early() fails.
> > The purpose is to make sure that klp_init_*_early()
> > will actually never fail.
> >
> > This might make things somehow complicated. Any future change
> > in klp_init_*_early() might require change in pre-early-init()
> > to catch eventual problems earlier.
>
> I'm not sure why that would be a problem. If we can separate out the
> things which fail from the things which don't, it simplifies things.
I think that there is no right answer. It depends on personal
preferences.
> > My proposal is to use simple trick. klp_free_patch_*() iterate
> > using the dynamic list (list_for_each_entry) while klp_init_*_early()
> > iterate using the arrays.
> >
> > Now, we just need to make sure that klp_init_*_early() will only add
> > fully initialized structures into the dynamic list. As a result,
> > klp_free_patch() will see only the fully initialized structures
> > and could release them. It will not see that not-yet-initialized
> > structures but it is fine. They are not initialized and they do not
> > need to be freed.
> >
> > In fact, I think that klp_init_*_early() functions already do
> > the right thing. IMHO, if you move the module_get() then you
> > could safely do:
> >
> > int klp_enable_patch(struct klp_patch *patch)
> > {
> > [...]
> > if (!try_module_get(patch->mod)) {
> > mutex_unlock(&klp_mutex);
> > return -ENODEV;
> > }
> >
> > ret = klp_init_patch_early(patch);
> > if (ret)
> > goto err;
> >
> >
> > Note that it would need to get tested.
> >
> > Anyway, does it make sense now?
>
> It would work, but it's too clever/non-obvious. If we rely on tricks
> then we may eventually forget about them and accidentally break
> assumptions later.
It is not that super magic from my POV. But again, it is a personal
taste.
I do not want to bikeshed about it. Feel free to use the pre-early
thing.
Best Regards,
Petr
prev parent reply other threads:[~2021-12-20 9:48 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-12-14 22:01 [PATCH v2] livepatch: Fix leak on klp_init_patch_early failure path David Vernet
2021-12-14 23:51 ` Josh Poimboeuf
2021-12-15 10:06 ` Petr Mladek
2021-12-15 15:20 ` David Vernet
2021-12-17 13:51 ` Petr Mladek
2021-12-17 21:50 ` Josh Poimboeuf
2021-12-20 9:48 ` Petr Mladek [this message]
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=YcBReVgF2hFLpyjf@alley \
--to=pmladek@suse.com \
--cc=corbet@lwn.net \
--cc=gregkh@linuxfoundation.org \
--cc=jikos@kernel.org \
--cc=joe.lawrence@redhat.com \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=live-patching@vger.kernel.org \
--cc=mbenes@suse.cz \
--cc=songliubraving@fb.com \
--cc=void@manifault.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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).