DPDK patches and discussions
 help / color / mirror / Atom feed
From: Akhil Goyal <gakhil@marvell.com>
To: Akhil Goyal <gakhil@marvell.com>,
	Ciara Power <ciara.power@intel.com>,
	"dev@dpdk.org" <dev@dpdk.org>
Cc: "roy.fan.zhang@intel.com" <roy.fan.zhang@intel.com>,
	"declan.doherty@intel.com" <declan.doherty@intel.com>,
	"stable@dpdk.org" <stable@dpdk.org>,
	ZhihongX Peng <zhihongx.peng@intel.com>,
	Anoob Joseph <anoobj@marvell.com>
Subject: Re: [dpdk-dev] [EXT] [PATCH] crypto: fix heap use after free bug
Date: Fri, 30 Jul 2021 19:10:23 +0000	[thread overview]
Message-ID: <CO6PR18MB44849460E0AC88113F729262D8EC9@CO6PR18MB4484.namprd18.prod.outlook.com> (raw)
In-Reply-To: <PH0PR18MB4491A7059B135724136C7B96D8E99@PH0PR18MB4491.namprd18.prod.outlook.com>

Fixed title
Cryptodev: fix heap use after free
> > The PMD destroy function was calling the release function, which frees
> > cryptodev->data, and then tries to free cryptodev->data->dev_private,
> > which causes the heap use after free issue.
> >
> > A temporary pointer is set before the free of cryptodev->data,
> > which can then be used afterwards to free dev_private.
> > The free cannot be moved to before the release function is called,
> > as dev_private is used in the QAT close function while being released.
I believe all PMDs use dev_private for close.
Hence replaces QAT with PMD
> >
> > Fixes: 9e6edea41805 ("cryptodev: add APIs to assist PMD initialisation")
> > Cc: declan.doherty@intel.com
> > Cc: stable@dpdk.org
> >
> > Reported-by: ZhihongX Peng <zhihongx.peng@intel.com>
> > Signed-off-by: Ciara Power <ciara.power@intel.com>
> >
> > ---
> > The same issue is found in crypto/octeontx,
> > which may need to be addressed by maintainers.
> > Cc: Anoob Joseph <anoobj@marvell.com>
> > ---
> >  lib/cryptodev/rte_cryptodev_pmd.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/lib/cryptodev/rte_cryptodev_pmd.c
> > b/lib/cryptodev/rte_cryptodev_pmd.c
> > index 0912004127..900acd7ba4 100644
> > --- a/lib/cryptodev/rte_cryptodev_pmd.c
> > +++ b/lib/cryptodev/rte_cryptodev_pmd.c
> > @@ -140,6 +140,7 @@ int
> >  rte_cryptodev_pmd_destroy(struct rte_cryptodev *cryptodev)
> >  {
> >  	int retval;
> > +	void *tmp_dev_private = cryptodev->data->dev_private;
> 
> Can we rename this pointer as dev_private?

Renamed this while merging, as we have RC3 deadline today.
> 
> >
> >  	CDEV_LOG_INFO("Closing crypto device %s", cryptodev->device-
> > >name);
> >
> > @@ -149,7 +150,7 @@ rte_cryptodev_pmd_destroy(struct rte_cryptodev
> > *cryptodev)
> >  		return retval;
> >
> >  	if (rte_eal_process_type() == RTE_PROC_PRIMARY)
> > -		rte_free(cryptodev->data->dev_private);
> > +		rte_free(tmp_dev_private);
> >
> >
> >  	cryptodev->device = NULL;
> > --
> > 2.25.1


  reply	other threads:[~2021-07-30 19:10 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-21 12:51 [dpdk-dev] " Ciara Power
2021-07-27 18:04 ` [dpdk-dev] [EXT] " Akhil Goyal
2021-07-30 19:10   ` Akhil Goyal [this message]
2021-07-30 19:11     ` Akhil Goyal

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CO6PR18MB44849460E0AC88113F729262D8EC9@CO6PR18MB4484.namprd18.prod.outlook.com \
    --to=gakhil@marvell.com \
    --cc=anoobj@marvell.com \
    --cc=ciara.power@intel.com \
    --cc=declan.doherty@intel.com \
    --cc=dev@dpdk.org \
    --cc=roy.fan.zhang@intel.com \
    --cc=stable@dpdk.org \
    --cc=zhihongx.peng@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).