linux-kernel - Re: [PATCH 2/2] SUNRPC: Optimize 'svc_print

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

Message-ID: <87r1xe7pvy.fsf@notabene.neil.brown.name>
Date:   Fri, 27 Mar 2020 08:44:17 +1100
From:   NeilBrown <neilb@...e.de>
To:     Christophe JAILLET <christophe.jaillet@...adoo.fr>,
        linux-kernel@...r.kernel.org
Cc:     linux-nfs@...r.kernel.org, kernel-janitors@...r.kernel.org,
        kernel-janitors@...r.kernel.org, linux-kernel@...r.kernel.org,
        kernel-janitors@...r.kernel.org, linux-kernel@...r.kernel.org,
        kernel-janitors@...r.kernel.org
Subject: Re: [PATCH 2/2] SUNRPC: Optimize 'svc_print_xprts()'

On Thu, Mar 26 2020, Christophe JAILLET wrote:

> Le 25/03/2020 à 23:53, NeilBrown a écrit :
>> Can I suggest something more like this:
>> diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
>> index de3c077733a7..0292f45b70f6 100644
>> --- a/net/sunrpc/svc_xprt.c
>> +++ b/net/sunrpc/svc_xprt.c
>> @@ -115,16 +115,9 @@ int svc_print_xprts(char *buf, int maxlen)
>>   	buf[0] = '\0';
>>   
>>   	spin_lock(&svc_xprt_class_lock);
>> -	list_for_each_entry(xcl, &svc_xprt_class_list, xcl_list) {
>> -		int slen;
>> -
>> -		sprintf(tmpstr, "%s %d\n", xcl->xcl_name, xcl->xcl_max_payload);
>> -		slen = strlen(tmpstr);
>> -		if (len + slen > maxlen)
>> -			break;
>> -		len += slen;
>> -		strcat(buf, tmpstr);
>> -	}
>> +	list_for_each_entry(xcl, &svc_xprt_class_list, xcl_list)
>> +		len += scnprintf(buf + len, maxlen - len, "%s %d\n",
>> +				 xcl->xcl_name, xcl->xcl_max_payload);
>>   	spin_unlock(&svc_xprt_class_lock);
>>   
>>   	return len;
>>
>> NeilBrown
>
> Hi,
>
> this was what I suggested in the patch:
>      ---
>      This patch should have no functional change.
>      We could go further, use scnprintf and write directly in the 
> destination
>      buffer. However, this could lead to a truncated last line.
>      ---

Sorry - I missed that.
So add

 end = strrchr(tmpstr, '\n');
 if (end)
    end[1] = 0;
 else
    tmpstr[0] = 0;

or maybe something like
	list_for_each_entry(xcl, &svc_xprt_class_list, xcl_list) {
		int l = snprintf(buf + len, maxlen - len, "%s %d\n",
				 xcl->xcl_name, xcl->xcl_max_payload);
                if (l < maxlen - len)
                	len += l;
        }
        buf[len] = 0;

There really is no need to have the secondary buffer, and I think doing
so just complicates the code.
That last version is a change of behaviour in that it will skip over
lines that are too long, rather than aborting on the first one.
I don't know which is preferred.

Thanks,
NeilBrown
 

>
> And Chuck Lever confirmed that:
>      That's exactly what this function is trying to avoid. As part of any
>      change in this area, it would be good to replace the current block
>      comment before this function with a Doxygen-format comment that
>      documents that goal.
>
> So, I will only send a V2 based on comments already received.
>
> CJ

Download attachment "signature.asc" of type "application/pgp-signature" (833 bytes)