linux-kernel - Re: [PATCH v4 1/5] exfat: integrates dir-entry getting and validation

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 27 Aug 2020 21:26:21 +0900
From:   Tetsuhiro Kohada <kohada.t2@...il.com>
To:     Namjae Jeon <namjae.jeon@...sung.com>
Cc:     kohada.tetsuhiro@...mitsubishielectric.co.jp,
        mori.takahiro@...mitsubishielectric.co.jp,
        motai.hirotaka@...mitsubishielectric.co.jp,
        'Sungjong Seo' <sj1557.seo@...sung.com>,
        linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 1/5] exfat: integrates dir-entry getting and validation

Thank you for your quick review.

On 2020/08/27 12:19, Namjae Jeon wrote:
>> +	i = ES_INDEX_NAME;
>> +	while ((ep = exfat_get_validated_dentry(es, i++, TYPE_NAME))) {
> Please find the way to access name entries like ep_file, ep_stream
> without calling exfat_get_validated_dentry().

Hmm, it's a hard order.
I can't separate length/type validation and extraction.
Sorry, I have no good idea.


>> @@ -590,17 +587,16 @@ int exfat_remove_entries(struct inode *inode, struct exfat_chain *p_dir,  void
>> exfat_update_dir_chksum_with_entry_set(struct exfat_entry_set_cache *es)  {
>>   	int chksum_type = CS_DIR_ENTRY, i;
>> -	unsigned short chksum = 0;
>> +	u16 chksum = 0;
>>   	struct exfat_dentry *ep;
>>
>>   	for (i = 0; i < es->num_entries; i++) {
>> -		ep = exfat_get_dentry_cached(es, i);
>> +		ep = exfat_get_validated_dentry(es, i, TYPE_ALL);
> Ditto, You do not need to repeatedly call exfat_get_validated_dentry() for the entries
> which got from exfat_get_dentry_set().

Even if I could do that, it would be very difficult to implement a checksum patch.
It is also difficult to use for rename, move, delete.
(these also have no verification of neme-length and set-checksum)


>>   	/* validiate cached dentries */
>> -	for (i = 1; i < num_entries; i++) {
>> -		ep = exfat_get_dentry_cached(es, i);
>> -		if (!exfat_validate_entry(exfat_get_entry_type(ep), &mode))
>> -			goto free_es;
>> +	es->ep_stream = exfat_get_validated_dentry(es, ES_INDEX_STREAM, TYPE_STREAM);
>> +	if (!es->ep_stream)
>> +		goto free_es;
>> +
>> +	if (max_entries == ES_ALL_ENTRIES) {
>> +		for (i = 0; i < ES_FILE(es).num_ext; i++)
>> +			if (!exfat_get_validated_dentry(es, ES_INDEX_STREAM + i, TYPE_SECONDARY))
>> +				goto free_es;
>> +		for (i = 0; i * EXFAT_FILE_NAME_LEN < ES_STREAM(es).name_len; i++)
>> +			if (!exfat_get_validated_dentry(es, ES_INDEX_NAME + i, TYPE_NAME))
>> +				goto free_es;
> Why do you unnecessarily check entries with two loops?
> Please refer to the patch I sent.

This order is possible.
However, TYPE_SECONDARY loop will be back as checksum loop.

In the next patch, I can fix the 'TYPE_SECONDARY loop' order.
do you need it?


BR
---
Tetsuhiro Kohada <kohada.t2@...il.com>