linux-kernel - Re: [PATCH] kvm: fix coalesced

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]

Date:	Wed, 27 May 2009 16:32:33 -0400
From:	Gregory Haskins <gregory.haskins@...il.com>
To:	Chris Wright <chrisw@...s-sol.org>
CC:	Gregory Haskins <ghaskins@...ell.com>, avi@...hat.com,
	kvm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] kvm: fix coalesced_mmio leak on shutdown

Chris Wright wrote:
> * Gregory Haskins (ghaskins@...ell.com) wrote:
>   
>> It would appear that we are invoking kfree() on the wrong pointer in the
>> destructor for the coalesced_mmio device.  This would result in a potential
>> leak during shutdown.
>>     
>
> Happens to work and not leak:
>
> struct kvm_coalesced_mmio_dev { 
>         struct kvm_io_device dev;
>         struct kvm *kvm;
>         int nb_zones;
>         struct kvm_coalesced_mmio_zone zone[KVM_COALESCED_MMIO_ZONE_MAX];
> };
>
>   
Ah, yes.  That explains it.  Still sloppy, tho.

>> Signed-off-by: Gregory Haskins <ghaskins@...ell.com>
>> ---
>>
>>  virt/kvm/coalesced_mmio.c |    5 ++++-
>>  1 files changed, 4 insertions(+), 1 deletions(-)
>>
>> diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c
>> index 5ae620d..03ea280 100644
>> --- a/virt/kvm/coalesced_mmio.c
>> +++ b/virt/kvm/coalesced_mmio.c
>> @@ -80,7 +80,10 @@ static void coalesced_mmio_write(struct kvm_io_device *this,
>>  
>>  static void coalesced_mmio_destructor(struct kvm_io_device *this)
>>  {
>> -	kfree(this);
>> +	struct kvm_coalesced_mmio_dev *dev =
>> +		(struct kvm_coalesced_mmio_dev *)this->private;
>>     
>
> I think container_of() makes more sense here.
>   

I was working on that patch when I noticed the "leak" above.  Figured I
should send the fix out first, in case my container_of patch is shot down.

Just polishing it up now.  Will send out soon.

-Greg


Download attachment "signature.asc" of type "application/pgp-signature" (267 bytes)