linux-kernel - Re: [Qemu-devel] [PATCH] fw_cfg: use __ATTR_RO

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: <20190227142651.3enr2eey5bn2v6ls@master>
Date:   Wed, 27 Feb 2019 14:26:51 +0000
From:   Wei Yang <richard.weiyang@...il.com>
To:     "Michael S. Tsirkin" <mst@...hat.com>
Cc:     Wei Yang <richardw.yang@...ux.intel.com>, somlo@....edu,
        qemu-devel@...gnu.org, linux-kernel@...r.kernel.org
Subject: Re: [Qemu-devel] [PATCH] fw_cfg: use __ATTR_RO_MODE to define rev
 sysfs

On Wed, Feb 27, 2019 at 08:51:11AM -0500, Michael S. Tsirkin wrote:
>On Wed, Feb 27, 2019 at 01:33:19PM +0800, Wei Yang wrote:
>> On Tue, Feb 26, 2019 at 11:10:06AM -0500, Michael S. Tsirkin wrote:
>> >On Tue, Feb 26, 2019 at 03:31:59PM +0800, Wei Yang wrote:
>> >> Leverage __ATTR_RO_MODE to define rev sysfs instead of using open code
>> >> to define the attribute.
>> >> 
>> >> Signed-off-by: Wei Yang <richardw.yang@...ux.intel.com>
>> >> ---
>> >>  drivers/firmware/qemu_fw_cfg.c | 13 ++++---------
>> >>  1 file changed, 4 insertions(+), 9 deletions(-)
>> >> 
>> >> diff --git a/drivers/firmware/qemu_fw_cfg.c b/drivers/firmware/qemu_fw_cfg.c
>> >> index 039e0f91dba8..a1293cbd7adb 100644
>> >> --- a/drivers/firmware/qemu_fw_cfg.c
>> >> +++ b/drivers/firmware/qemu_fw_cfg.c
>> >> @@ -296,18 +296,13 @@ static int fw_cfg_do_platform_probe(struct platform_device *pdev)
>> >>  	return 0;
>> >>  }
>> >>  
>> >> -static ssize_t fw_cfg_showrev(struct kobject *k, struct attribute *a, char *buf)
>> >> +static ssize_t fw_cfg_rev_show(struct kobject *k, struct kobj_attribute *a,
>> >> +			       char *buf)
>> >>  {
>> >>  	return sprintf(buf, "%u\n", fw_cfg_rev);
>> >>  }
>> >> -
>> >> -static const struct {
>> >> -	struct attribute attr;
>> >> -	ssize_t (*show)(struct kobject *k, struct attribute *a, char *buf);
>> >> -} fw_cfg_rev_attr = {
>> >> -	.attr = { .name = "rev", .mode = S_IRUSR },
>> >> -	.show = fw_cfg_showrev,
>> >> -};
>> >> +static const struct kobj_attribute fw_cfg_rev_attr =
>> >> +	__ATTR_RO_MODE(fw_cfg_rev, 0400);
>> >>  
>> >>  /* fw_cfg_sysfs_entry type */
>> >>  struct fw_cfg_sysfs_entry {
>> >
>> >
>> >Looks like this will change the name from "rev" to "fw_cfg_rev".
>> >That's a userspace visible change which we should not do lightly.
>> 
>> You are right, I should keep the interface untouched.
>> 
>> To keep it user un-visible, we could change like below:
>> 
>> -       __ATTR_RO(fw_cfg_rev);
>> +       __ATTR_RO(rev);
>> 
>> Is this better for you?
>
>
>Also why use 0400 and not S_IRUSR?
>

This is interesting. The scripts/checkpatch.pl suggest to use 0400.

I am not sure why the script give this suggestion. Maybe we need to fix
the script?

>> >> -- 
>> >> 2.19.1
>> 
>> -- 
>> Wei Yang
>> Help you, Help me

-- 
Wei Yang
Help you, Help me