[PATCH] KEYS: Don't write out to userspace while holding key semaphore

[Waiman Long]

On 3/5/20 4:16 PM, Eric Biggers wrote:
> On Thu, Mar 05, 2020 at 04:06:40PM -0500, Waiman Long wrote:
>> diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c
>> index 9b898c969558..564a4d187329 100644
>> --- a/security/keys/keyctl.c
>> +++ b/security/keys/keyctl.c
>> @@ -846,14 +846,36 @@ long keyctl_read_key(key_serial_t keyid, char __user *buffer, size_t buflen)
>>  can_read_key:
>>  	ret = -EOPNOTSUPP;
>>  	if (key->type->read) {
>> -		/* Read the data with the semaphore held (since we might sleep)
>> +		/*
>> +		 * Read the data with the semaphore held (since we might sleep)
>>  		 * to protect against the key being updated or revoked.
>> +		 *
>> +		 * Allocating a temporary buffer to hold the keys before
>> +		 * transferring them to user buffer to avoid potential
>> +		 * deadlock involving page fault and mmap_sem.
>>  		 */
>> +		char *tmpbuf = kmalloc(buflen, GFP_KERNEL);
> This is passing an arbitrarily large size from userspace into kmalloc().
>
> - Eric
>
That is true. Is there a limit of how much key data are expected to be
returned from the kernel? We could set an internal limit here on how
large the buffer can be.

Cheers,
Longman