linux-kernel - Re: [PATCH-next] lib: parser: optimize match

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: <8ed91647-f429-493e-b4e1-1f6bf11c228b@huawei.com>
Date:   Sat, 10 Dec 2022 10:51:11 +0800
From:   "lilingfeng (A)" <lilingfeng3@...wei.com>
To:     Tejun Heo <tj@...nel.org>
CC:     <linux-kernel@...r.kernel.org>, <jack@...e.cz>,
        <bingjingc@...ology.com>, <ebiggers@...gle.com>,
        <james.smart@...adcom.com>, <houtao1@...wei.com>,
        <yi.zhang@...wei.com>, <yangerkun@...wei.com>, <yukuai3@...wei.com>
Subject: Re: [PATCH-next] lib: parser: optimize match_NUMER apis to use local
 array


在 2022/12/10 0:57, Tejun Heo 写道:
> Hello,
>
> In general, I think this is a great idea. Some nits below:
>
> On Fri, Dec 09, 2022 at 02:34:34PM +0800, Li Lingfeng wrote:
>> +/*
>> + * max size needed by diffrent bases to express U64
>> + * HEX: "0xFFFFFFFFFFFFFFFF" --> 18
>> + * DEC: "18446744073709551615" --> 20
>> + * OCT: "01777777777777777777777" --> 23
>> + * pick the max one to define U64_MAX_SIZE
>> + */
>> +#define U64_MAX_SIZE 23
> Bikeshedding but how about naming it like NUMBER_BUF_LEN and including the
> space for '\0'? Or just give it some extra space and make it 32 bytes.
Yes, it's my mistake, I'll send a new patch soon.
>> +static int match_strdup_local(const substring_t *s, char *buf)
> I find it weird to name this as generic as match_strdup_local() and make it
> assume that the buffer length is U64_MAX_SIZE + 1. Maybe just let the caller
> pass in the buffer length as a parameter? Then, it's just strcpy and there
> already is match_strlcpy() so we don't need this at all.

Thank you for your advice. But I think match_number() is aimed to turn the
string to num, so maybe it's better to return an error code rather than
using match_stlcpy() to truncate it to give a wrong num when the string
is too long to store.

>> +{
>> +	size_t len = s->to - s->from;
>> +
>> +	if (!s->from)
>> +		return -EINVAL;
> If we use match_strlcpy() we lose the above null check but given that other
> match_*() functions aren't doing it, this likely shouldn't matter.

Like this:
match_strdup
  kmemdup_nul
   if (!s) // null check has been done here
    return NULL
So I think null check may be necessary.

Thanks.

> Thanks.
>