linux-kernel - Re: [PATCH] nvmem: fix unregistering device in nvmem

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: <9e94f0fd-e2d5-4d9e-5759-a5f591191785@gmail.com>
Date:   Tue, 21 Dec 2021 18:46:01 +0100
From:   Rafał Miłecki <zajec5@...il.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     Srinivas Kandagatla <srinivas.kandagatla@...aro.org>,
        Johan Hovold <johan@...nel.org>,
        Andrey Smirnov <andrew.smirnov@...il.com>,
        linux-kernel@...r.kernel.org,
        Rafał Miłecki <rafal@...ecki.pl>
Subject: Re: [PATCH] nvmem: fix unregistering device in nvmem_register() error
 path

On 21.12.2021 17:06, Greg Kroah-Hartman wrote:
> On Tue, Dec 21, 2021 at 04:45:50PM +0100, Rafał Miłecki wrote:
>> From: Rafał Miłecki <rafal@...ecki.pl>
>>
>> 1. Drop incorrect put_device() calls
>>
>> If device_register() fails then underlaying device_add() takes care of
>> calling put_device() if needed. There is no need to do that in a driver.
> 
> Did you read the documentation for device_register() that says:
> 
>   * NOTE: _Never_ directly free @dev after calling this function, even
>   * if it returned an error! Always use put_device() to give up the
>   * reference initialized in this function instead.

I clearly tried to be too smart and ignored documentation.

I'd say device_add() behaviour is rather uncommon and a bit unintuitive.
Most kernel functions are safe to assume to do nothing that requires
cleanup if they fail.

E.g. if I call platform_device_register() and it fails I don't need to
call anything like platform_device_put(). I just free previously
allocated memory.

When calling device_register() / device_add() it seems device always
gets partially registered (even if it fails!). Enough to make it safe to
depend on core subsystem calling .release() after device_put().

So what initially looks like unbalanced device_put() call is actually
some device_add() specific magic behaviour ;)

Sorry. I should have checked documentation before posting patches.
That's not my best day.


>> 2. Use device_unregister()
>>
>> Now that we don't call put_device() we can use above helper.
>>
>> Fixes: 3360acdf8391 ("nvmem: core: fix leaks on registration errors")
>> Cc: Johan Hovold <johan@...nel.org>
>> Signed-off-by: Rafał Miłecki <rafal@...ecki.pl>
>> ---
>> That put_device() was explicitly added by Johan but after checking
>> device_register() twice I still think it's incorrect. I hope I didn't
>> miss sth obvious and I didn't mess it up.
>> ---
>>   drivers/nvmem/core.c | 10 ++++------
>>   1 file changed, 4 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/nvmem/core.c b/drivers/nvmem/core.c
>> index 785a56e33f69..f7f31af7226f 100644
>> --- a/drivers/nvmem/core.c
>> +++ b/drivers/nvmem/core.c
>> @@ -901,12 +901,12 @@ struct nvmem_device *nvmem_register(const struct nvmem_config *config)
>>   
>>   	rval = device_register(&nvmem->dev);
>>   	if (rval)
>> -		goto err_put_device;
>> +		return ERR_PTR(rval);
> 
> Where do you call put_device() to free the allocated memory?
> 
> You just leaked the kzalloc() call to allocate the memory pointed to by
> nvmem :(
> 
> I think the code is fine as-is.

Yeah, I forgot about:

ida_free(&nvmem_ida, nvmem->id);
kfree(nvmem);