linux-kernel - Re: [PATCH] RDMA/ocrdma: fix use after free in ocrdma_dealloc_ucontext

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:   Thu, 7 Jan 2021 13:43:53 -0800
From:   Tom Rix <trix@...hat.com>
To:     Jason Gunthorpe <jgg@...dia.com>
Cc:     selvin.xavier@...adcom.com, devesh.sharma@...adcom.com,
        dledford@...hat.com, leon@...nel.org, maxg@...lanox.com,
        galpress@...zon.com, michaelgur@...dia.com, monis@...lanox.com,
        gustavoars@...nel.org, linux-rdma@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] RDMA/ocrdma: fix use after free in
 ocrdma_dealloc_ucontext_pd()


On 1/7/21 12:41 PM, Jason Gunthorpe wrote:
> On Tue, Dec 29, 2020 at 06:46:53PM -0800, trix@...hat.com wrote:
>> From: Tom Rix <trix@...hat.com>
>>
>> In ocrdma_dealloc_ucontext_pd() uctx->cntxt_pd is assigned to
>> the variable pd and then after uctx->cntxt_pd is freed, the
>> variable pd is passed to function _ocrdma_dealloc_pd() which
>> dereferences pd directly or through its call to
>> ocrdma_mbx_dealloc_pd().
>>
>> Reorder the free using the variable pd.
>>
>> Fixes: 21a428a019c9 ("RDMA: Handle PD allocations by IB/core")
>> Signed-off-by: Tom Rix <trix@...hat.com>
>>  drivers/infiniband/hw/ocrdma/ocrdma_verbs.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
> Applied to for-rc
>
> Is anyone testing ocrdma? Just doing the pyverbs rdma tests with kasn
> turned on would have instantly caught this, and the change is nearly a
> year old.
>
> Is ocrdma obsolete enough we can delete the driver?

I am not an authority on ocrdma, i am fixing treewide, the problems clang static analysis flags.

Tom

>
> Thanks,
> Jason
>