Linux-Fsdevel Archive on lore.kernel.org
help / color / mirror / Atom feed
From: "Namjae Jeon" <namjae.jeon@samsung.com>
To: "'Tetsuhiro Kohada'" <kohada.t2@gmail.com>
Cc: <kohada.tetsuhiro@dc.mitsubishielectric.co.jp>,
	<mori.takahiro@ab.mitsubishielectric.co.jp>,
	<motai.hirotaka@aj.mitsubishielectric.co.jp>,
	"'Sungjong Seo'" <sj1557.seo@samsung.com>,
	<linux-fsdevel@vger.kernel.org>, <linux-kernel@vger.kernel.org>
Subject: RE: [PATCH 1/2] exfat: add NameLength check when extracting name
Date: Thu, 13 Aug 2020 11:53:06 +0900	[thread overview]
Message-ID: <001001d6711c$def48c80$9cdda580$@samsung.com> (raw)
In-Reply-To: <ed561c17-3b85-9bf1-e765-5d9a15786585@gmail.com>

> Thank you for your reply.
> 
> >> -static void exfat_get_uniname_from_ext_entry(struct super_block *sb,
> >> -		struct exfat_chain *p_dir, int entry, unsigned short *uniname)
> >> +static int exfat_get_uniname_from_name_entries(struct exfat_entry_set_cache *es,
> >> +		struct exfat_uni_name *uniname)
> >>   {
> >> -	int i;
> >> -	struct exfat_entry_set_cache *es;
> >> +	int n, l, i;
> >>   	struct exfat_dentry *ep;
> >>
> >> -	es = exfat_get_dentry_set(sb, p_dir, entry, ES_ALL_ENTRIES);
> >> -	if (!es)
> >> -		return;
> >> +	uniname->name_len = es->de_stream->name_len;
> >> +	if (uniname->name_len == 0)
> >> +		return -EIO;
> > Can we validate ->name_len and name entry ->type in exfat_get_dentry_set() ?
> 
> Yes.
> As I wrote in a previous email, entry type validation, name-length validation, and name extraction
> should not be separated, so implement all of these in exfat_get_dentry_set().
> It can be easily implemented by adding uniname to exfat_entry_set_cache and calling
> exfat_get_uniname_from_name_entries() from exfat_get_dentry_set().
No, We can check stream->name_len and name entry type in exfat_get_dentry_set().
And you are already checking entry type with TYPE_SECONDARY in
exfat_get_dentry_set(). Why do we have to check twice?

>
> However, that would be over-implementation.
> Not all callers of exfat_get_dentry_set() need a name.
Where? It will not checked with ES_2_ENTRIES.

> It is enough to validate the name when it is needed.
> This is a file-system driver, not fsck.
Sorry, I don't understand what you are talking about. If there is a problem
in ondisk-metadata, Filesystem should return error.

> Validation is possible in exfat_get_dentry_set(), but unnecessary.
> 
> Why do you want to validate the name in exfat_get_dentry_set()?
exfat_get_dentry_set validates file, stream entry. And you are trying to check
name entries with type_secondary. In addition, trying add the checksum check.
Conversely, Why would you want to add those checks to exfat_get_dentry_set()?
Why do we check only name entries separately? Aren't you intent to return
validated entry set in exfat_get_dentry_set()?
> 
> 
> BR
> ---
> Tetsuhiro Kohada <kohada.t2@gmail.com>


  reply	other threads:[~2020-08-13  2:53 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20200806055718epcas1p33009b21ebf96b48d6e3f819065fade28@epcas1p3.samsung.com>
2020-08-06  5:56 ` Tetsuhiro Kohada
2020-08-06  5:56   ` [PATCH 2/2] exfat: unify name extraction Tetsuhiro Kohada
2020-08-08 17:19     ` Sungjong Seo
2020-08-12  6:02       ` Tetsuhiro Kohada
2020-08-21 10:41         ` Sungjong Seo
2020-08-25 10:15           ` Tetsuhiro Kohada
2020-08-08 16:54   ` [PATCH 1/2] exfat: add NameLength check when extracting name Sungjong Seo
2020-08-12  4:53     ` Tetsuhiro Kohada
2020-08-10  6:13   ` Namjae Jeon
2020-08-12 15:04     ` Tetsuhiro Kohada
2020-08-13  2:53       ` Namjae Jeon [this message]
2020-08-17  9:26         ` Kohada.Tetsuhiro

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='001001d6711c$def48c80$9cdda580$@samsung.com' \
    --to=namjae.jeon@samsung.com \
    --cc=kohada.t2@gmail.com \
    --cc=kohada.tetsuhiro@dc.mitsubishielectric.co.jp \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mori.takahiro@ab.mitsubishielectric.co.jp \
    --cc=motai.hirotaka@aj.mitsubishielectric.co.jp \
    --cc=sj1557.seo@samsung.com \
    --subject='RE: [PATCH 1/2] exfat: add NameLength check when extracting name' \
    /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

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 NNTP newsgroup(s).