Received-SPF: pass (google.com: domain of linux-kernel+bounces-153082-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Date: Mon, 22 Apr 2024 11:20:15 +0200
From: Lukas Wunner <lukas@wunner.de>
To: Dan Williams <dan.j.williams@intel.com>
Cc: gregkh@linuxfoundation.org,
	Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>,
	Marc Herbert <marc.herbert@intel.com>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
	linux-coco@lists.linux.dev, alsa-devel@alsa-project.org
Subject: Re: [PATCH 1/3] sysfs: Fix crash on empty group attributes array
Message-ID: <ZiYrzzk9Me1aksmE@wunner.de>
References: <170863444851.1479840.10249410842428140526.stgit@dwillia2-xfh.jf.intel.com>
 <170863445442.1479840.1818801787239831650.stgit@dwillia2-xfh.jf.intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <170863445442.1479840.1818801787239831650.stgit@dwillia2-xfh.jf.intel.com>

On Thu, Feb 22, 2024 at 12:40:54PM -0800, Dan Williams wrote:
> It turns out that arch/x86/events/intel/core.c makes use of "empty"
> attributes.
> 
> 	static struct attribute *empty_attrs;
> 
> 	__init int intel_pmu_init(void)
> 	{
> 	        struct attribute **extra_skl_attr = &empty_attrs;
> 	        struct attribute **extra_attr = &empty_attrs;
> 	        struct attribute **td_attr    = &empty_attrs;
> 	        struct attribute **mem_attr   = &empty_attrs;
> 	        struct attribute **tsx_attr   = &empty_attrs;
> 		...
> 
> That breaks the assumption __first_visible() that expects that if
> grp->attrs is set then grp->attrs[0] must also be set and results in
> backtraces like:
> 
>     BUG: kernel NULL pointer dereference, address: 00rnel mode
>     #PF: error_code(0x0000) - not-present ] PREEMPT SMP NOPTI
>     CPU: 1 PID: 1 Comm: swapper/IP: 0010:exra_is_visible+0x14/0x20
>      ? exc_page_fault+0x68/0x190
>      internal_create_groups+0x42/0xa0
>      pmu_dev_alloc+0xc0/0xe0
>      perf_event_sysfs_init+0x580000000000 ]---
>     RIP: 0010:exra_is_visible+0x14/0
> 
> Check for non-empty attributes array before calling is_visible().
[...]
> --- a/fs/sysfs/group.c
> +++ b/fs/sysfs/group.c
> @@ -33,10 +33,10 @@ static void remove_files(struct kernfs_node *parent,
>  
>  static umode_t __first_visible(const struct attribute_group *grp, struct kobject *kobj)
>  {
> -	if (grp->attrs && grp->is_visible)
> +	if (grp->attrs && grp->attrs[0] && grp->is_visible)
>  		return grp->is_visible(kobj, grp->attrs[0], 0);
>  
> -	if (grp->bin_attrs && grp->is_bin_visible)
> +	if (grp->bin_attrs && grp->bin_attrs[0] && grp->is_bin_visible)
>  		return grp->is_bin_visible(kobj, grp->bin_attrs[0], 0);
>  
>  	return 0;

I'm wondering why 0 is returned by default and not SYSFS_GROUP_INVISIBLE.

An empty attribute list (containing just the NULL sentinel) will now
result in the attribute group being visible as an empty directory.

I thought the whole point was to hide such empty directories.

Was it a conscious decision to return 0?
Did you expect breakage if SYSFS_GROUP_INVISIBLE is returned?

Thanks,

Lukas