linux-kernel - Re: [PATCH 3/6] edac, mce_amd_inj: Modify flags attrigute to use string arguments

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]

Message-ID: <556CAF9F.6080008@amd.com>
Date:	Mon, 1 Jun 2015 14:16:47 -0500
From:	Aravind Gopalakrishnan <aravind.gopalakrishnan@....com>
To:	Borislav Petkov <bp@...en8.de>
CC:	<dougthompson@...ssion.com>, <mchehab@....samsung.com>,
	<linux-edac@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 3/6] edac, mce_amd_inj: Modify flags attrigute to use
 string arguments

On 5/29/2015 8:49 AM, Borislav Petkov wrote:
> On Wed, May 27, 2015 at 02:03:35PM -0500, Aravind Gopalakrishnan wrote:
>>   
>> +#define MAX_FLAG_OPT_SIZE	10
> Why 10?
>
> This should be 2 and increased when another, longer injection type
> string gets introduced.
>

So I hit an issue when I have this as 2 and moving 'ret = cnt' statement 
(in flags_write() below) to after the check for MAX_FLAG_OPT_SIZE.

The problem was, if I did echo hw > flags;
I would get an error on dmesg like so-
[   78.692949] flags_write: Invalid flags value:

But the write would have actually gone through. i.e, if we do cat flags, 
the output was 'hw'.

The issue seems to be that 'cnt' of flags_write() is 3 (after accounting 
for a terminating NULL) when we enter the function (for this case),

Since we move 'ret = cnt;' statement to after the check, we end up 
returning 2.
And since it is less than 'cnt', we re-enter flags_write() for the final 
NULL character.
At this point, our comparisons with flags_options[] fail and we end up 
returning -EINVAL from __set_inj().
Hence the error on dmesg..

The fix to this was simply having MAX_FLAG_OPT_SIZE as 3. In this case, 
we already account for the NULL and flags_write()
returns the correct value upon success.
So I shall go ahead and make that change to MAX_FLAG_OPT_SIZE as 3 and 
increase it in subsequent patch for the apic interrupts.

Thanks,
-Aravind.

>>   
>> -DEFINE_SIMPLE_ATTRIBUTE(flags_fops, flags_get, flags_set, "%llu\n");
>> +static ssize_t flags_write(struct file *filp, const char __user *ubuf,
>> +			   size_t cnt, loff_t *ppos)
>> +{
>> +	char buf[MAX_FLAG_OPT_SIZE + 1];
>> +	int err;
>> +	size_t ret;
>> +
>> +	ret = cnt;
> You're assigning cnt to ret here...
>
>> +
>> +	if (cnt > MAX_FLAG_OPT_SIZE)
>> +		cnt = MAX_FLAG_OPT_SIZE;
> ... but correcting cnt afterwards. The assignment should be *after* that
> correction.
>
>> +
>> +	if (copy_from_user(&buf, ubuf, cnt))
>> +		return -EFAULT;
>> +
>> +	buf[cnt] = 0;
>> +
>> +	/* strip whitespaces.. */
>> +	strstrip(buf);
>> +
>> +	err = __set_inj(buf, cnt - 1);
>> +	if (err) {
>> +		pr_err("%s: Invalid flags value: %s\n", __func__, buf);
>> +		return err;
>> +	}
>> +
>> +	*ppos += ret;
>> +
>> +	return ret;
>> +}
>> +
>>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/