* Re: [PATCH rdma-rc 0/9] mlx4 fixes for 4.9
From: Leon Romanovsky @ 2016-11-08 8:50 UTC (permalink / raw)
To: dledford-H+wXaHxf7aLQT0dZR+AlfA, Yuval Shaia
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1478375842-21513-1-git-send-email-leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 159 bytes --]
On Sat, Nov 05, 2016 at 09:57:13PM +0200, Leon Romanovsky wrote:
> Hi Doug,
Please drop this patch series, I'll respin it once I'll return to
office.
Thanks
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply
* Re: FW: [Bug 185551] New: rxe_rdma: RTNL: assertion failed at net/core/ethtool.c (550)
From: Yonatan Cohen @ 2016-11-08 9:41 UTC (permalink / raw)
To: v.badalyan-mR0gyJWGnJgvJsYlp49lxw; +Cc: linux-rdma
In-Reply-To: <HE1PR05MB157874BECC0B171C75933E1BA7A40-eBadYZ65MZ/9T9KkpKWrANqRiQSDpxhJvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
On 11/6/2016 1:08 PM, Majd Dibbiny wrote:
>
> https://bugzilla.kernel.org/show_bug.cgi?id=185551
>
> Bug ID: 185551
> Summary: rxe_rdma: RTNL: assertion failed at net/core/ethtool.c
> (550)
> Product: Drivers
> Version: 2.5
> Kernel Version: 4.8.5-1.el7.elrepo.x86_64
> Hardware: All
> OS: Linux
> Tree: Mainline
> Status: NEW
> Severity: high
> Priority: P1
> Component: Infiniband/RDMA
> Assignee: drivers_infiniband-rdma-ztI5WcYan/vQLgFONoPN62D2FQJk+8+b@public.gmane.org
> Reporter: v.badalyan-mR0gyJWGnJgvJsYlp49lxw@public.gmane.org
> Regression: No
>
> #rxe_cfg start
> #rxe_cfg add eno1.4004
> #rxe_cfg
>
> [root@intel2-i5 ~]# rxe_cfg
> Name Link Driver Speed NMTU IPv4_addr RDEV RMTU
> DevNet yes bridge
> eno1 yes e1000e
> eno1.4003 yes 802.1Q
> eno1.4004 yes 802.1Q rxe0 4096 (5)
> enp2s1 yes r8169
> ovirtmgmt yes bridge
>
>
how do i reproduce this ?
was rxe device coupled with a bridge-interface ?
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: SQ overflow seen running isert traffic
From: Potnuri Bharat Teja @ 2016-11-08 10:06 UTC (permalink / raw)
To: Nicholas A. Bellinger
Cc: SWise OGC, 'Sagi Grimberg',
target-devel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
In-Reply-To: <1477885208.27946.8.camel-XoQW25Eq2zviZyQQd+hFbcojREIfoBdhmpATvIKMPHk@public.gmane.org>
On Monday, October 10/31/16, 2016 at 09:10:08 +0530, Nicholas A. Bellinger wrote:
> Hi Steve, Potnuri, & Co,
>
> On Tue, 2016-10-18 at 09:34 -0500, Steve Wise wrote:
> > >
> > > > I tried out this change and it works fine with iwarp. I dont see SQ
> > > > overflow. Apparently we have increased the sq too big to overflow. I am
> > going
> > > > to let it run with higher workloads for longer time, to see if it holds
> > good.
> > >
> > > Actually on second thought, this patch is an overkill. Effectively we
> > > now set:
> > >
> > > MAX_CMD=266
> > > and max_rdma_ctx=128 so together we take 394 which seems to too much.
> > >
> > > If we go by the scheme of 1 rdma + 1 send for each IO we need:
> > > - 128 sends
> > > - 128 rdmas
> > > - 10 miscs
> > >
> > > so this gives 266.
> > >
> > > Perhaps this is due to the fact that iWARP needs to register memory for
> > > rdma reads as well? (and also rdma writes > 128k for chelsio HW right?)
> > >
> >
> > iWARP definitely needs to register memory for the target of reads, due to
> > REMOTE_WRITE requirement for the protocol. The source of a write doesn't need
> > to register memory, but the SGE depth can cause multiple WRITE WRs to be
> > required to service the IO. And in theory there should be some threshold where
> > it might be better performance-wise to do a memory register + 1 WRITE vs X
> > WRITEs.
> >
> > As you mentioned, the RW API should account for this, but perhaps it is still
> > off some. Bharat, have a look into the RDMA-RW API and let us see if we can
> > figure out if the additional SQ depth it adds is sufficient.
> >
> > > What is the workload you are running? with immediatedata enabled you
> > > should issue reg+rdma_read+send only for writes > 8k.
> > >
> > > Does this happen when you run only reads for example?
> > >
> > > I guess its time to get the sq accounting into shape...
> >
> > So to sum up - 2 issues:
> >
> > 1) we believe the iSER + RW API correctly sizes the SQ, yet we're seeing SQ
> > overflows. So the SQ sizing needs more investigation.
> >
> > 2) if the SQ is full, then the iSER/target code is supposed to resubmit. And
> > apparently that isn't working.
> >
>
> For #2, target-core expects -ENOMEM or -EAGAIN return from fabric driver
> callbacks to signal internal queue-full retry logic. Otherwise, the
> extra se_cmd->cmd_kref response SCF_ACK_KREF is leaked until session
> shutdown and/or reinstatement occurs.
>
> AFAICT, Potunri's earlier hung task with v4.8.y + ABORT_TASK is likely
> the earlier v4.1+ regression:
>
> https://github.com/torvalds/linux/commit/527268df31e57cf2b6d417198717c6d6afdb1e3e
>
> That said, there is room for improvement in target-core queue-full error
> signaling, and iscsi-target/iser-target callback error propagation.
>
> Sending out a series shortly to address these particular items.
> Please have a look.
>
Thanks for the changes Nicholas.
Testing them now.
Thanks,
Bharat.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH rdma-core 1/7] libhns: Add initial main frame
From: Leon Romanovsky @ 2016-11-08 12:54 UTC (permalink / raw)
To: Jason Gunthorpe
Cc: oulijun, dledford-H+wXaHxf7aLQT0dZR+AlfA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA,
linuxarm-hv44wF8Li93QT0dZR+AlfA
In-Reply-To: <20161107231532.GB7002-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 797 bytes --]
On Mon, Nov 07, 2016 at 04:15:32PM -0700, Jason Gunthorpe wrote:
> On Sat, Oct 29, 2016 at 09:16:25AM +0800, oulijun wrote:
>
> > We hope that the only one userspace library file named
> > libhns-rdmav2.so will be used for the different hardware
> > version(hip06, hip07, ...), because there are only little change
> > between their userspace drivers. So we need to distinguish hardware
> > version.
>
> I guess that makes sense, but you still need to be able to parse dt
> compatible strings that are lists.
IMHO, it can be easily done as follow up patches.
>
> Jason
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply
* Re: [RESEND REQEUST] consult with the libhns upstreaming into rdma-core
From: Leon Romanovsky @ 2016-11-08 12:56 UTC (permalink / raw)
To: oulijun; +Cc: Doug Ledford, linux-rdma, Linuxarm
In-Reply-To: <58202286.3020303-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 583 bytes --]
On Mon, Nov 07, 2016 at 02:43:18PM +0800, oulijun wrote:
> Hi Jason/Doug,
> Following your comments with I have fixed several issues.
> Thank you for pointing them out.
> Do I need to resend the PATCH v2 for libhns?
No, there is no need to resend it.
All patches are trapped by patchworks and we will grab them from there.
Thanks
>
> Lijun Ou
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply
* Re: rdma-core example spec file is broken
From: Alaa Hleihel @ 2016-11-08 14:47 UTC (permalink / raw)
To: Jason Gunthorpe
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA, leonro-VPRAkNaXOzVWk0Htik3J/w,
yishaih-VPRAkNaXOzVWk0Htik3J/w
In-Reply-To: <20161107164415.GA8780-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
Hi
On 11/7/2016 18:44, Jason Gunthorpe wrote:
> On Sun, Nov 06, 2016 at 02:01:20PM +0200, Alaa Hleihel wrote:
>> Hi Jason,
>>
>> The example spec file is broken after commit ddd530be4622 ("Minor RPM spec file improvments").
>>
>> rpmbuild fails with the following error:
>>
>> Executing(%build): /bin/sh -e /var/tmp/rpm-tmp.7l0yoN
>> + umask 022
>> + cd /tmp/xxx/BUILD
>> + cd rdma-core-11
>> /var/tmp/rpm-tmp.7l0yoN: line 31: syntax error near unexpected token `post'
>> error: Bad exit status from /var/tmp/rpm-tmp.7l0yoN (%build)
>>
>> The issue is that the new "Requires" macros were added after the %prep and %build sections..
> Ah, interesting, in my test scripts the common spec file is
> superceeded by the version in redhat/ for systemd systems, which is
> why I did not notice it.
Yes, it happens with relatively new versions only..
> Jason
>
> From f2388d9d110ab43295d50d7345d5dda289a87a41 Mon Sep 17 00:00:00 2001
> From: Jason Gunthorpe <jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
> Date: Mon, 7 Nov 2016 09:37:38 -0700
> Subject: [PATCH] Fix common rdma-core.spec for systemd systems
>
> RPM needs the Requires lines for systemd to be earlier.
>
> Signed-off-by: Jason Gunthorpe <jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
> ---
> rdma-core.spec | 10 +++++++---
> 1 file changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/rdma-core.spec b/rdma-core.spec
> index 35186c3271b8..bea2a2267d1d 100644
> --- a/rdma-core.spec
> +++ b/rdma-core.spec
> @@ -46,6 +46,13 @@ BuildRequires: make
> %endif
> %endif
>
> +# Detect if systemd is supported on this system
> +%if 0%{?_unitdir:1}
> +Requires(post): systemd-units
> +Requires(preun): systemd-units
> +Requires(postun): systemd-units
> +%endif
> +
> %description
> Temporary packaging
>
> @@ -59,9 +66,6 @@ This is a simple example without the split sub packages to get things started.
> # Detect if systemd is supported on this system
> %if 0%{?_unitdir:1}
> %define my_unitdir %{_unitdir}
> -Requires(post): systemd-units
> -Requires(preun): systemd-units
> -Requires(postun): systemd-units
> %else
> %define my_unitdir /tmp/
> %endif
I tested this patch.
It resolves the issue.
Thanks,
Alaa
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* RE: [PATCH rdma-next 0/4] Add packet pacing support for IB verbs
From: Hefty, Sean @ 2016-11-08 17:49 UTC (permalink / raw)
To: Leon Romanovsky, dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
In-Reply-To: <1477909297-14491-1-git-send-email-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
> When sending from a 10G host to a 1G host, it is easy to overrun the
> receiver,
> leading to packet loss and traffic backing off. Similar problems occur
> when
> a 10G host sends data to a sub-10G virtual circuit, or a 40G host
> sending
> to a 10G host. Packet pacing could control packet injection rate and
> reduces
> network congestion to maximize throughput & minimize network latency.
Why isn't the path record data and existing mechanisms sufficient to handle this?
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* RE: nvmet_rdma crash - DISCONNECT event with NULL queue
From: Steve Wise @ 2016-11-08 20:45 UTC (permalink / raw)
To: 'Sagi Grimberg', 'Christoph Hellwig'
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA,
linux-nvme-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r
In-Reply-To: <b499abb0-afc9-2985-c4c4-3ceba4ca6f33-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org>
> > Running the same test over mlx4/roce, I hit a warning in list_debug, and
then a
> > stuck CPU...
>
> Steve,
>
> Does this patch help?
> --
Hey Sagi,
With this patch applied, I haven't seen any more target side crashes. I say
this with a caveat, though, that I'm chasing a few other host side issues
still...
Steve.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* [PATCH v4 0/9] SELinux support for Infiniband RDMA
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw, paul, sds, eparis, dledford, sean.hefty, hal.rosenstock
Cc: selinux, linux-security-module, linux-rdma, yevgenyp, liranl,
leonro, Daniel Jurgens
From: Daniel Jurgens <danielj@mellanox.com>
Infiniband applications access HW from user-space -- traffic is generated
directly by HW, bypassing the kernel. Consequently, Infiniband Partitions,
which are associated directly with HW transport endpoints, are a natural
choice for enforcing granular mandatory access control for Infiniband. QPs may
only send or receives packets tagged with the corresponding partition key
(PKey). The PKey is not a cryptographic key; it's a 16 bit number identifying
the partition.
Every Infiniband fabric is controlled by a central Subnet Manager (SM). The SM
provisions the partitions by assigning each port with the partitions it can
access. In addition, the SM tags each port with a subnet prefix, which
identifies the subnet. Determining which users are allowed to access which
partition keys on a given subnet forms an effective policy for isolating users
on the fabric. Any application that attempts to send traffic on a given subnet
is automatically subject to the policy, regardless of which device and port it
uses. SM software configures the subnet through a privileged Subnet Management
Interface (SMI), which is presented by each Infiniband port. Thus, the SMI must
also be controlled to prevent unauthorized changes to fabric configuration and
partitioning.
To support access control for IB partitions and subnet management, security
contexts must be provided for two new types of objects - PKeys and IB ports.
A PKey label consists of a subnet prefix and a range of PKey values and is
similar to the labeling mechanism for netports. Each Infiniband port can reside
on a different subnet. So labeling the PKey values for specific subnet prefixes
provides the user maximum flexibility, as PKey values may be determined
independently for different subnets. There is a single access vector for PKeys
called "access".
An Infiniband port is labeled by device name and port number. There is a single
access vector for IB ports called "manage_subnet".
Because RDMA allows kernel bypass, enforcement must be done during connection
setup. Communication over RDMA requires a send and receive queue, collectively
known as a Queue Pair (QP). A QP must be initialized by privileged system calls
before it can be used to send or receive data. During initialization the user
must provide the PKey and port the QP will use; at this time access control can
be enforced.
Because there is a possibility that the enforcement settings or security
policy can change, a means of notifying the ib_core module of such changes
is required. To facilitate this a generic notification callback mechanism
is added to the LSM. One callback is registered for checking the QP PKey
associations when the policy changes. Mad agents also register a callback,
they cache the permission to send and receive SMPs to avoid another per
packet call to the LSM.
Because frequent accesses to the same PKey's SID is expected a cache is
implemented which is very similar to the netport cache.
In order to properly enforce security when changes to the PKey table or
security policy or enforcement occur ib_core must track which QPs are
using which port, pkey index, and alternate path for every IB device.
This makes operations that used to be atomic transactional.
When modifying a QP, ib_core must associate it with the PKey index, port,
and alternate path specified. If the QP was already associated with
different settings, the QP is added to the new list prior to the
modification. If the modify succeeds then the old listing is removed. If
the modify fails the new listing is removed and the old listing remains
unchanged.
When destroying a QP the ib_qp structure is freed by the decive specific
driver (i.e. mlx4_ib) if the 'destroy' is successful. This requires storing
security related information in a separate structure. When a 'destroy'
request is in process the ib_qp structure is in an undefined state so if
there are changes to the security policy or PKey table, the security checks
cannot reset the QP if it doesn't have permission for the new setting. If
the 'destroy' fails, security for that QP must be enforced again and its
status in the list is restored. If the 'destroy' succeeds the security info
can be cleaned up and freed.
There are a number of locks required to protect the QP security structure
and the QP to device/port/pkey index lists. If multiple locks are required,
the safe locking order is: QP security structure mutex first, followed by
any list locks needed, which are sorted first by port followed by pkey
index.
---
v2:
- Use void* blobs in the LSM hooks. Paul Moore
- Make the policy change callback generic. Yuval Shaia, Paul Moore
- Squash LSM changes into the patches where the calls are added. Paul Moore
- Don't add new initial SIDs. Stephen Smalley
- Squash MAD agent PKey and SMI patches and move logic to IB security. Dan Jurgens
- Changed ib_end_port to ib_port. Paul Moore
- Changed ib_port access vector from smp to manage_subnet. Paul Moore
- Added pkey and ib_port details to the audit log. Paul Moore
- See individual patches for more detail.
v3:
- ib_port -> ib_endport. Paul Moore
- use notifier chains for LSM notifications. Paul Moore
- reorder parameters in hooks to put security blob first. Paul Moore
- Don't treat device name as untrusted string in audit log. Paul Moore
v4:
- Added seperate AVC callback for LSM notifier. Paul Moore
- Removed unneeded braces in ocontext_read. Paul Moore
Daniel Jurgens (9):
IB/core: IB cache enhancements to support Infiniband security
IB/core: Enforce PKey security on QPs
selinux lsm IB/core: Implement LSM notification system
IB/core: Enforce security on management datagrams
selinux: Create policydb version for Infiniband support
selinux: Allocate and free infiniband security hooks
selinux: Implement Infiniband PKey "Access" access vector
selinux: Add IB Port SMP access vector
selinux: Add a cache for quicker retreival of PKey SIDs
drivers/infiniband/core/Makefile | 3 +-
drivers/infiniband/core/cache.c | 57 ++-
drivers/infiniband/core/core_priv.h | 115 ++++++
drivers/infiniband/core/device.c | 86 +++++
drivers/infiniband/core/mad.c | 52 ++-
drivers/infiniband/core/security.c | 709 +++++++++++++++++++++++++++++++++++
drivers/infiniband/core/uverbs_cmd.c | 20 +-
drivers/infiniband/core/verbs.c | 27 +-
include/linux/lsm_audit.h | 15 +
include/linux/lsm_hooks.h | 35 ++
include/linux/security.h | 35 ++
include/rdma/ib_mad.h | 4 +
include/rdma/ib_verbs.h | 49 +++
security/Kconfig | 9 +
security/lsm_audit.c | 18 +
security/security.c | 59 +++
security/selinux/Makefile | 2 +-
security/selinux/hooks.c | 86 ++++-
security/selinux/ibpkey.c | 245 ++++++++++++
security/selinux/include/classmap.h | 4 +
security/selinux/include/ibpkey.h | 31 ++
security/selinux/include/objsec.h | 11 +
security/selinux/include/security.h | 7 +-
security/selinux/selinuxfs.c | 2 +
security/selinux/ss/policydb.c | 129 ++++++-
security/selinux/ss/policydb.h | 27 +-
security/selinux/ss/services.c | 83 ++++
27 files changed, 1875 insertions(+), 45 deletions(-)
create mode 100644 drivers/infiniband/core/security.c
create mode 100644 security/selinux/ibpkey.c
create mode 100644 security/selinux/include/ibpkey.h
--
1.8.3.1
^ permalink raw reply
* [PATCH v4 1/9] IB/core: IB cache enhancements to support Infiniband security
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw, paul, sds, eparis, dledford, sean.hefty, hal.rosenstock
Cc: selinux, linux-security-module, linux-rdma, yevgenyp, liranl,
leonro, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj@mellanox.com>
From: Daniel Jurgens <danielj@mellanox.com>
Cache the subnet prefix and add a function to access it. Enforcing
security requires frequent queries of the subnet prefix and the pkeys in
the pkey table.
Also removed an unneded pr_warn about memory allocation failure.
issue: 736423
Change-Id: Ifdef64b097a8d1d55db65f08ce401d9d2e4b025e
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Reviewed-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
---
v2:
- In ib_get_cached_subnet_prefix wait to initialize p until after
validation. Yuval Shaia
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
---
drivers/infiniband/core/cache.c | 36 ++++++++++++++++++++++++++++++++++--
drivers/infiniband/core/core_priv.h | 3 +++
include/rdma/ib_verbs.h | 1 +
3 files changed, 38 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c
index 1a2984c..affc8ef 100644
--- a/drivers/infiniband/core/cache.c
+++ b/drivers/infiniband/core/cache.c
@@ -934,6 +934,26 @@ int ib_get_cached_pkey(struct ib_device *device,
}
EXPORT_SYMBOL(ib_get_cached_pkey);
+int ib_get_cached_subnet_prefix(struct ib_device *device,
+ u8 port_num,
+ u64 *sn_pfx)
+{
+ unsigned long flags;
+ int p;
+
+ if (port_num < rdma_start_port(device) ||
+ port_num > rdma_end_port(device))
+ return -EINVAL;
+
+ p = port_num - rdma_start_port(device);
+ read_lock_irqsave(&device->cache.lock, flags);
+ *sn_pfx = device->cache.subnet_prefix_cache[p];
+ read_unlock_irqrestore(&device->cache.lock, flags);
+
+ return 0;
+}
+EXPORT_SYMBOL(ib_get_cached_subnet_prefix);
+
int ib_find_cached_pkey(struct ib_device *device,
u8 port_num,
u16 pkey,
@@ -1110,6 +1130,8 @@ static void ib_cache_update(struct ib_device *device,
device->cache.lmc_cache[port - rdma_start_port(device)] = tprops->lmc;
+ device->cache.subnet_prefix_cache[port - rdma_start_port(device)] =
+ tprops->subnet_prefix;
write_unlock_irq(&device->cache.lock);
kfree(gid_cache);
@@ -1168,9 +1190,18 @@ int ib_cache_setup_one(struct ib_device *device)
(rdma_end_port(device) -
rdma_start_port(device) + 1),
GFP_KERNEL);
+
+ device->cache.subnet_prefix_cache =
+ kcalloc((rdma_end_port(device) - rdma_start_port(device) + 1),
+ sizeof(*device->cache.subnet_prefix_cache),
+ GFP_KERNEL);
+
if (!device->cache.pkey_cache ||
- !device->cache.lmc_cache) {
- pr_warn("Couldn't allocate cache for %s\n", device->name);
+ !device->cache.lmc_cache ||
+ !device->cache.subnet_prefix_cache) {
+ kfree(device->cache.pkey_cache);
+ kfree(device->cache.lmc_cache);
+ kfree(device->cache.subnet_prefix_cache);
return -ENOMEM;
}
@@ -1213,6 +1244,7 @@ void ib_cache_release_one(struct ib_device *device)
gid_table_release_one(device);
kfree(device->cache.pkey_cache);
kfree(device->cache.lmc_cache);
+ kfree(device->cache.subnet_prefix_cache);
}
void ib_cache_cleanup_one(struct ib_device *device)
diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h
index 19d499d..ce826e4 100644
--- a/drivers/infiniband/core/core_priv.h
+++ b/drivers/infiniband/core/core_priv.h
@@ -153,4 +153,7 @@ int ib_nl_handle_set_timeout(struct sk_buff *skb,
int ib_nl_handle_ip_res_resp(struct sk_buff *skb,
struct netlink_callback *cb);
+int ib_get_cached_subnet_prefix(struct ib_device *device,
+ u8 port_num,
+ u64 *sn_pfx);
#endif /* _CORE_PRIV_H */
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 5ad43a4..db178fd 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -1761,6 +1761,7 @@ struct ib_cache {
struct ib_pkey_cache **pkey_cache;
struct ib_gid_table **gid_cache;
u8 *lmc_cache;
+ u64 *subnet_prefix_cache;
};
struct ib_dma_mapping_ops {
--
1.8.3.1
^ permalink raw reply related
* [PATCH v4 2/9] IB/core: Enforce PKey security on QPs
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw, paul, sds, eparis, dledford, sean.hefty, hal.rosenstock
Cc: selinux, linux-security-module, linux-rdma, yevgenyp, liranl,
leonro, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj@mellanox.com>
From: Daniel Jurgens <danielj@mellanox.com>
Add new LSM hooks to allocate and free security contexts and check for
permission to access a PKey.
Allocate and free a security context when creating and destroying a QP.
This context is used for controlling access to PKeys.
When a request is made to modify a QP that changes the port, PKey index,
or alternate path, check that the QP has permission for the PKey in the
PKey table index on the subnet prefix of the port. If the QP is shared
make sure all handles to the QP also have access.
Store which port and PKey index a QP is using. After the reset to init
transition the user can modify the port, PKey index and alternate path
independently. So port and PKey settings changes can be a merge of the
previous settings and the new ones.
In order to maintain access control if there are PKey table or subnet
prefix change keep a list of all QPs are using each PKey index on
each port. If a change occurs all QPs using that device and port must
have access enforced for the new cache settings.
These changes add a transaction to the QP modify process. Association
with the old port and PKey index must be maintained if the modify fails,
and must be removed if it succeeds. Association with the new port and
PKey index must be established prior to the modify and removed if the
modify fails.
1. When a QP is modified to a particular Port, PKey index or alternate
path insert that QP into the appropriate lists.
2. Check permission to access the new settings.
3. If step 2 grants access attempt to modify the QP.
4a. If steps 2 and 3 succeed remove any prior associations.
4b. If ether fails remove the new setting associations.
If a PKey table or subnet prefix changes walk the list of QPs and
check that they have permission. If not send the QP to the error state
and raise a fatal error event. If it's a shared QP make sure all the
QPs that share the real_qp have permission as well. If the QP that
owns a security structure is denied access the security structure is
marked as such and the QP is added to an error_list. Once the moving
the QP to error is complete the security structure mark is cleared.
Maintaining the lists correctly turns QP destroy into a transaction.
The hardware driver for the device frees the ib_qp structure, so while
the destroy is in progress the ib_qp pointer in the ib_qp_security
struct is undefined. When the destroy process begins the ib_qp_security
structure is marked as destroying. This prevents any action from being
taken on the QP pointer. After the QP is destroyed successfully it
could still listed on an error_list wait for it to be processed by that
flow before cleaning up the structure.
If the destroy fails the QPs port and PKey settings are reinserted into
the appropriate lists, the destroying flag is cleared, and access control
is enforced, in case there were any cache changes during the destroy
flow.
To keep the security changes isolated a new file is used to hold security
related functionality.
issue: 736423
Change-Id: I0509141a57a8e08c91a9e0c059800251949b18e6
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
---
v2:
- Squashed LSM hook additions. Paul Moore
- Changed security blobs to void*. Paul Moore
v3:
- Change parameter order of pkey_access hook. Paul Moore
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
---
drivers/infiniband/core/Makefile | 3 +-
drivers/infiniband/core/cache.c | 21 +-
drivers/infiniband/core/core_priv.h | 77 +++++
drivers/infiniband/core/device.c | 33 ++
drivers/infiniband/core/security.c | 617 +++++++++++++++++++++++++++++++++++
drivers/infiniband/core/uverbs_cmd.c | 20 +-
drivers/infiniband/core/verbs.c | 27 +-
include/linux/lsm_hooks.h | 27 ++
include/linux/security.h | 21 ++
include/rdma/ib_verbs.h | 48 +++
security/Kconfig | 9 +
security/security.c | 31 ++
12 files changed, 925 insertions(+), 9 deletions(-)
create mode 100644 drivers/infiniband/core/security.c
diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile
index edaae9f..da4e2c1 100644
--- a/drivers/infiniband/core/Makefile
+++ b/drivers/infiniband/core/Makefile
@@ -10,7 +10,8 @@ obj-$(CONFIG_INFINIBAND_USER_ACCESS) += ib_uverbs.o ib_ucm.o \
ib_core-y := packer.o ud_header.o verbs.o cq.o rw.o sysfs.o \
device.o fmr_pool.o cache.o netlink.o \
roce_gid_mgmt.o mr_pool.o addr.o sa_query.o \
- multicast.o mad.o smi.o agent.o mad_rmpp.o
+ multicast.o mad.o smi.o agent.o mad_rmpp.o \
+ security.o
ib_core-$(CONFIG_INFINIBAND_USER_MEM) += umem.o
ib_core-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += umem_odp.o umem_rbtree.o
diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c
index affc8ef..48eaeca 100644
--- a/drivers/infiniband/core/cache.c
+++ b/drivers/infiniband/core/cache.c
@@ -53,6 +53,7 @@ struct ib_update_work {
struct work_struct work;
struct ib_device *device;
u8 port_num;
+ bool enforce_security;
};
union ib_gid zgid;
@@ -1046,7 +1047,8 @@ int ib_get_cached_lmc(struct ib_device *device,
EXPORT_SYMBOL(ib_get_cached_lmc);
static void ib_cache_update(struct ib_device *device,
- u8 port)
+ u8 port,
+ bool enforce_security)
{
struct ib_port_attr *tprops = NULL;
struct ib_pkey_cache *pkey_cache = NULL, *old_pkey_cache;
@@ -1134,6 +1136,11 @@ static void ib_cache_update(struct ib_device *device,
tprops->subnet_prefix;
write_unlock_irq(&device->cache.lock);
+ if (enforce_security)
+ ib_security_cache_change(device,
+ port,
+ tprops->subnet_prefix);
+
kfree(gid_cache);
kfree(old_pkey_cache);
kfree(tprops);
@@ -1150,7 +1157,9 @@ static void ib_cache_task(struct work_struct *_work)
struct ib_update_work *work =
container_of(_work, struct ib_update_work, work);
- ib_cache_update(work->device, work->port_num);
+ ib_cache_update(work->device,
+ work->port_num,
+ work->enforce_security);
kfree(work);
}
@@ -1171,6 +1180,12 @@ static void ib_cache_event(struct ib_event_handler *handler,
INIT_WORK(&work->work, ib_cache_task);
work->device = event->device;
work->port_num = event->element.port_num;
+ if (event->event == IB_EVENT_PKEY_CHANGE ||
+ event->event == IB_EVENT_GID_CHANGE)
+ work->enforce_security = true;
+ else
+ work->enforce_security = false;
+
queue_work(ib_wq, &work->work);
}
}
@@ -1211,7 +1226,7 @@ int ib_cache_setup_one(struct ib_device *device)
return err;
for (p = 0; p <= rdma_end_port(device) - rdma_start_port(device); ++p)
- ib_cache_update(device, p + rdma_start_port(device));
+ ib_cache_update(device, p + rdma_start_port(device), true);
INIT_IB_EVENT_HANDLER(&device->cache.event_handler,
device, ib_cache_event);
diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h
index ce826e4..68e3de0 100644
--- a/drivers/infiniband/core/core_priv.h
+++ b/drivers/infiniband/core/core_priv.h
@@ -38,6 +38,14 @@
#include <rdma/ib_verbs.h>
+struct pkey_index_qp_list {
+ struct list_head pkey_index_list;
+ u16 pkey_index;
+ /* Lock to hold while iterating the qp_list. */
+ spinlock_t qp_list_lock;
+ struct list_head qp_list;
+};
+
#if IS_ENABLED(CONFIG_INFINIBAND_ADDR_TRANS_CONFIGFS)
int cma_configfs_init(void);
void cma_configfs_exit(void);
@@ -156,4 +164,73 @@ int ib_nl_handle_ip_res_resp(struct sk_buff *skb,
int ib_get_cached_subnet_prefix(struct ib_device *device,
u8 port_num,
u64 *sn_pfx);
+
+#ifdef CONFIG_SECURITY_INFINIBAND
+void ib_security_destroy_port_pkey_list(struct ib_device *device);
+
+void ib_security_cache_change(struct ib_device *device,
+ u8 port_num,
+ u64 subnet_prefix);
+
+int ib_security_modify_qp(struct ib_qp *qp,
+ struct ib_qp_attr *qp_attr,
+ int qp_attr_mask,
+ struct ib_udata *udata);
+
+int ib_create_qp_security(struct ib_qp *qp, struct ib_device *dev);
+void ib_destroy_qp_security_begin(struct ib_qp_security *sec);
+void ib_destroy_qp_security_abort(struct ib_qp_security *sec);
+void ib_destroy_qp_security_end(struct ib_qp_security *sec);
+int ib_open_shared_qp_security(struct ib_qp *qp, struct ib_device *dev);
+void ib_close_shared_qp_security(struct ib_qp_security *sec);
+#else
+static inline void ib_security_destroy_port_pkey_list(struct ib_device *device)
+{
+}
+
+static inline void ib_security_cache_change(struct ib_device *device,
+ u8 port_num,
+ u64 subnet_prefix)
+{
+}
+
+static inline int ib_security_modify_qp(struct ib_qp *qp,
+ struct ib_qp_attr *qp_attr,
+ int qp_attr_mask,
+ struct ib_udata *udata)
+{
+ return qp->device->modify_qp(qp->real_qp,
+ qp_attr,
+ qp_attr_mask,
+ udata);
+}
+
+static inline int ib_create_qp_security(struct ib_qp *qp,
+ struct ib_device *dev)
+{
+ return 0;
+}
+
+static inline void ib_destroy_qp_security_begin(struct ib_qp_security *sec)
+{
+}
+
+static inline void ib_destroy_qp_security_abort(struct ib_qp_security *sec)
+{
+}
+
+static inline void ib_destroy_qp_security_end(struct ib_qp_security *sec)
+{
+}
+
+static inline int ib_open_shared_qp_security(struct ib_qp *qp,
+ struct ib_device *dev)
+{
+ return 0;
+}
+
+static inline void ib_close_shared_qp_security(struct ib_qp_security *sec)
+{
+}
+#endif
#endif /* _CORE_PRIV_H */
diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index 760ef60..5b42e83 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -320,6 +320,30 @@ void ib_get_device_fw_str(struct ib_device *dev, char *str, size_t str_len)
}
EXPORT_SYMBOL(ib_get_device_fw_str);
+static int setup_port_pkey_list(struct ib_device *device)
+{
+ int i;
+
+ /**
+ * device->port_pkey_list is indexed directly by the port number,
+ * Therefore it is declared as a 1 based array with potential empty
+ * slots at the beginning.
+ */
+ device->port_pkey_list = kzalloc(sizeof(*device->port_pkey_list)
+ * (rdma_end_port(device) + 1),
+ GFP_KERNEL);
+
+ if (!device->port_pkey_list)
+ return -ENOMEM;
+
+ for (i = 0; i < (rdma_end_port(device) + 1); i++) {
+ spin_lock_init(&device->port_pkey_list[i].list_lock);
+ INIT_LIST_HEAD(&device->port_pkey_list[i].pkey_list);
+ }
+
+ return 0;
+}
+
/**
* ib_register_device - Register an IB device with IB core
* @device:Device to register
@@ -357,6 +381,12 @@ int ib_register_device(struct ib_device *device,
goto out;
}
+ ret = setup_port_pkey_list(device);
+ if (ret) {
+ dev_warn(device->dma_device, "Couldn't create per port_pkey_list\n");
+ goto out;
+ }
+
ret = ib_cache_setup_one(device);
if (ret) {
pr_warn("Couldn't set up InfiniBand P_Key/GID cache\n");
@@ -427,6 +457,9 @@ void ib_unregister_device(struct ib_device *device)
ib_device_unregister_sysfs(device);
ib_cache_cleanup_one(device);
+ ib_security_destroy_port_pkey_list(device);
+ kfree(device->port_pkey_list);
+
down_write(&lists_rwsem);
spin_lock_irqsave(&device->client_data_lock, flags);
list_for_each_entry_safe(context, tmp, &device->client_data_list, list)
diff --git a/drivers/infiniband/core/security.c b/drivers/infiniband/core/security.c
new file mode 100644
index 0000000..45800dd
--- /dev/null
+++ b/drivers/infiniband/core/security.c
@@ -0,0 +1,617 @@
+/*
+ * Copyright (c) 2016 Mellanox Technologies Ltd. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifdef CONFIG_SECURITY_INFINIBAND
+
+#include <linux/security.h>
+#include <linux/completion.h>
+#include <linux/list.h>
+
+#include <rdma/ib_verbs.h>
+#include <rdma/ib_cache.h>
+#include "core_priv.h"
+
+static struct pkey_index_qp_list *get_pkey_idx_qp_list(struct ib_port_pkey *pp)
+{
+ struct pkey_index_qp_list *pkey = NULL;
+ struct pkey_index_qp_list *tmp_pkey;
+ struct ib_device *dev = pp->sec->dev;
+
+ spin_lock(&dev->port_pkey_list[pp->port_num].list_lock);
+ list_for_each_entry(tmp_pkey,
+ &dev->port_pkey_list[pp->port_num].pkey_list,
+ pkey_index_list) {
+ if (tmp_pkey->pkey_index == pp->pkey_index) {
+ pkey = tmp_pkey;
+ break;
+ }
+ }
+ spin_unlock(&dev->port_pkey_list[pp->port_num].list_lock);
+ return pkey;
+}
+
+static int get_pkey_and_subnet_prefix(struct ib_port_pkey *pp,
+ u16 *pkey,
+ u64 *subnet_prefix)
+{
+ struct ib_device *dev = pp->sec->dev;
+ int ret;
+
+ ret = ib_get_cached_pkey(dev, pp->port_num, pp->pkey_index, pkey);
+ if (ret)
+ return ret;
+
+ ret = ib_get_cached_subnet_prefix(dev, pp->port_num, subnet_prefix);
+
+ return ret;
+}
+
+static int enforce_qp_pkey_security(u16 pkey,
+ u64 subnet_prefix,
+ struct ib_qp_security *qp_sec)
+{
+ struct ib_qp_security *shared_qp_sec;
+ int ret;
+
+ ret = security_ib_pkey_access(qp_sec->security, subnet_prefix, pkey);
+ if (ret)
+ return ret;
+
+ if (qp_sec->qp == qp_sec->qp->real_qp) {
+ list_for_each_entry(shared_qp_sec,
+ &qp_sec->shared_qp_list,
+ shared_qp_list) {
+ ret = security_ib_pkey_access(shared_qp_sec->security,
+ subnet_prefix,
+ pkey);
+ if (ret)
+ return ret;
+ }
+ }
+ return 0;
+}
+
+/* The caller of this function must hold the QP security
+ * mutex of the QP of the security structure in *pps.
+ *
+ * It takes separate ports_pkeys and security structure
+ * because in some cases the pps will be for a new settings
+ * or the pps will be for the real QP and security structure
+ * will be for a shared QP.
+ */
+static int check_qp_port_pkey_settings(struct ib_ports_pkeys *pps,
+ struct ib_qp_security *sec)
+{
+ u64 subnet_prefix;
+ u16 pkey;
+ int ret = 0;
+
+ if (!pps)
+ return 0;
+
+ if (pps->main.state != IB_PORT_PKEY_NOT_VALID) {
+ get_pkey_and_subnet_prefix(&pps->main,
+ &pkey,
+ &subnet_prefix);
+
+ ret = enforce_qp_pkey_security(pkey,
+ subnet_prefix,
+ sec);
+ }
+ if (ret)
+ goto out;
+
+ if (pps->alt.state != IB_PORT_PKEY_NOT_VALID) {
+ get_pkey_and_subnet_prefix(&pps->alt,
+ &pkey,
+ &subnet_prefix);
+
+ ret = enforce_qp_pkey_security(pkey,
+ subnet_prefix,
+ sec);
+ }
+
+ if (ret)
+ goto out;
+
+out:
+ return ret;
+}
+
+/* The caller of this function must hold the QP security
+ * mutex.
+ */
+static void qp_to_error(struct ib_qp_security *sec)
+{
+ struct ib_qp_security *shared_qp_sec;
+ struct ib_qp_attr attr = {
+ .qp_state = IB_QPS_ERR
+ };
+ struct ib_event event = {
+ .event = IB_EVENT_QP_FATAL
+ };
+
+ /* If the QP is in the process of being destroyed
+ * the qp pointer in the security structure is
+ * undefined. It cannot be modified now.
+ */
+ if (sec->destroying)
+ return;
+
+ ib_modify_qp(sec->qp,
+ &attr,
+ IB_QP_STATE);
+
+ if (sec->qp->event_handler && sec->qp->qp_context) {
+ event.element.qp = sec->qp;
+ sec->qp->event_handler(&event,
+ sec->qp->qp_context);
+ }
+
+ list_for_each_entry(shared_qp_sec,
+ &sec->shared_qp_list,
+ shared_qp_list) {
+ struct ib_qp *qp = shared_qp_sec->qp;
+
+ if (qp->event_handler && qp->qp_context) {
+ event.element.qp = qp;
+ event.device = qp->device;
+ qp->event_handler(&event,
+ qp->qp_context);
+ }
+ }
+}
+
+static inline void check_pkey_qps(struct pkey_index_qp_list *pkey,
+ struct ib_device *device,
+ u8 port_num,
+ u64 subnet_prefix)
+{
+ struct ib_port_pkey *pp, *tmp_pp;
+ bool comp;
+ LIST_HEAD(to_error_list);
+ u16 pkey_val;
+
+ if (!ib_get_cached_pkey(device,
+ port_num,
+ pkey->pkey_index,
+ &pkey_val)) {
+ spin_lock(&pkey->qp_list_lock);
+ list_for_each_entry(pp, &pkey->qp_list, qp_list) {
+ if (atomic_read(&pp->sec->error_list_count))
+ continue;
+
+ if (enforce_qp_pkey_security(pkey_val,
+ subnet_prefix,
+ pp->sec)) {
+ atomic_inc(&pp->sec->error_list_count);
+ list_add(&pp->to_error_list,
+ &to_error_list);
+ }
+ }
+ spin_unlock(&pkey->qp_list_lock);
+ }
+
+ list_for_each_entry_safe(pp,
+ tmp_pp,
+ &to_error_list,
+ to_error_list) {
+ mutex_lock(&pp->sec->mutex);
+ qp_to_error(pp->sec);
+ list_del(&pp->to_error_list);
+ atomic_dec(&pp->sec->error_list_count);
+ comp = pp->sec->destroying;
+ mutex_unlock(&pp->sec->mutex);
+
+ if (comp)
+ complete(&pp->sec->error_complete);
+ }
+}
+
+/* The caller of this function must hold the QP security
+ * mutex.
+ */
+static int port_pkey_list_insert(struct ib_port_pkey *pp)
+{
+ struct pkey_index_qp_list *tmp_pkey;
+ struct pkey_index_qp_list *pkey;
+ struct ib_device *dev;
+ u8 port_num = pp->port_num;
+ int ret = 0;
+
+ if (pp->state != IB_PORT_PKEY_VALID)
+ return 0;
+
+ dev = pp->sec->dev;
+
+ pkey = get_pkey_idx_qp_list(pp);
+
+ if (!pkey) {
+ bool found = false;
+
+ pkey = kzalloc(sizeof(*pkey), GFP_KERNEL);
+ if (!pkey)
+ return -ENOMEM;
+
+ spin_lock(&dev->port_pkey_list[port_num].list_lock);
+ /* Check for the PKey again. A racing process may
+ * have created it.
+ */
+ list_for_each_entry(tmp_pkey,
+ &dev->port_pkey_list[port_num].pkey_list,
+ pkey_index_list) {
+ if (tmp_pkey->pkey_index == pp->pkey_index) {
+ kfree(pkey);
+ pkey = tmp_pkey;
+ found = true;
+ break;
+ }
+ }
+
+ if (!found) {
+ pkey->pkey_index = pp->pkey_index;
+ spin_lock_init(&pkey->qp_list_lock);
+ INIT_LIST_HEAD(&pkey->qp_list);
+ list_add(&pkey->pkey_index_list,
+ &dev->port_pkey_list[port_num].pkey_list);
+ }
+ spin_unlock(&dev->port_pkey_list[port_num].list_lock);
+ }
+
+ spin_lock(&pkey->qp_list_lock);
+ list_add(&pp->qp_list, &pkey->qp_list);
+ spin_unlock(&pkey->qp_list_lock);
+
+ pp->state = IB_PORT_PKEY_LISTED;
+
+ return ret;
+}
+
+/* The caller of this function must hold the QP security
+ * mutex.
+ */
+static void port_pkey_list_remove(struct ib_port_pkey *pp)
+{
+ struct pkey_index_qp_list *pkey;
+
+ if (pp->state != IB_PORT_PKEY_LISTED)
+ return;
+
+ pkey = get_pkey_idx_qp_list(pp);
+
+ spin_lock(&pkey->qp_list_lock);
+ list_del(&pp->qp_list);
+ spin_unlock(&pkey->qp_list_lock);
+
+ /* The setting may still be valid, i.e. after
+ * a destroy has failed for example.
+ */
+ pp->state = IB_PORT_PKEY_VALID;
+}
+
+static void destroy_qp_security(struct ib_qp_security *sec)
+{
+ security_ib_free_security(sec->security);
+ kfree(sec->ports_pkeys);
+ kfree(sec);
+}
+
+/* The caller of this function must hold the QP security
+ * mutex.
+ */
+static struct ib_ports_pkeys *get_new_pps(const struct ib_qp *qp,
+ const struct ib_qp_attr *qp_attr,
+ int qp_attr_mask)
+{
+ struct ib_ports_pkeys *new_pps;
+ struct ib_ports_pkeys *qp_pps = qp->qp_sec->ports_pkeys;
+
+ new_pps = kzalloc(sizeof(*new_pps), GFP_KERNEL);
+ if (!new_pps)
+ return NULL;
+
+ if (qp_attr_mask & (IB_QP_PKEY_INDEX | IB_QP_PORT)) {
+ if (!qp_pps) {
+ new_pps->main.port_num = qp_attr->port_num;
+ new_pps->main.pkey_index = qp_attr->pkey_index;
+ } else {
+ new_pps->main.port_num = (qp_attr_mask & IB_QP_PORT) ?
+ qp_attr->port_num :
+ qp_pps->main.port_num;
+
+ new_pps->main.pkey_index =
+ (qp_attr_mask & IB_QP_PKEY_INDEX) ?
+ qp_attr->pkey_index :
+ qp_pps->main.pkey_index;
+ }
+ new_pps->main.state = IB_PORT_PKEY_VALID;
+ } else if (qp_pps) {
+ new_pps->main.port_num = qp_pps->main.port_num;
+ new_pps->main.pkey_index = qp_pps->main.pkey_index;
+ if (qp_pps->main.state != IB_PORT_PKEY_NOT_VALID)
+ new_pps->main.state = IB_PORT_PKEY_VALID;
+ }
+
+ if (qp_attr_mask & IB_QP_ALT_PATH) {
+ new_pps->alt.port_num = qp_attr->alt_port_num;
+ new_pps->alt.pkey_index = qp_attr->alt_pkey_index;
+ new_pps->alt.state = IB_PORT_PKEY_VALID;
+ } else if (qp_pps) {
+ new_pps->alt.port_num = qp_pps->alt.port_num;
+ new_pps->alt.pkey_index = qp_pps->alt.pkey_index;
+ if (qp_pps->alt.state != IB_PORT_PKEY_NOT_VALID)
+ new_pps->alt.state = IB_PORT_PKEY_VALID;
+ }
+
+ new_pps->main.sec = qp->qp_sec;
+ new_pps->alt.sec = qp->qp_sec;
+ return new_pps;
+}
+
+int ib_open_shared_qp_security(struct ib_qp *qp, struct ib_device *dev)
+{
+ struct ib_qp *real_qp = qp->real_qp;
+ int ret;
+
+ ret = ib_create_qp_security(qp, dev);
+
+ if (ret)
+ goto out;
+
+ mutex_lock(&real_qp->qp_sec->mutex);
+ ret = check_qp_port_pkey_settings(real_qp->qp_sec->ports_pkeys,
+ qp->qp_sec);
+
+ if (ret)
+ goto ret;
+
+ if (qp != real_qp)
+ list_add(&qp->qp_sec->shared_qp_list,
+ &real_qp->qp_sec->shared_qp_list);
+ret:
+ mutex_unlock(&real_qp->qp_sec->mutex);
+ if (ret)
+ destroy_qp_security(qp->qp_sec);
+
+out:
+ return ret;
+}
+
+void ib_close_shared_qp_security(struct ib_qp_security *sec)
+{
+ struct ib_qp *real_qp = sec->qp->real_qp;
+
+ mutex_lock(&real_qp->qp_sec->mutex);
+ list_del(&sec->shared_qp_list);
+ mutex_unlock(&real_qp->qp_sec->mutex);
+
+ destroy_qp_security(sec);
+}
+
+int ib_create_qp_security(struct ib_qp *qp, struct ib_device *dev)
+{
+ int ret;
+
+ qp->qp_sec = kzalloc(sizeof(*qp->qp_sec), GFP_KERNEL);
+ if (!qp->qp_sec)
+ return -ENOMEM;
+
+ qp->qp_sec->qp = qp;
+ qp->qp_sec->dev = dev;
+ mutex_init(&qp->qp_sec->mutex);
+ INIT_LIST_HEAD(&qp->qp_sec->shared_qp_list);
+ atomic_set(&qp->qp_sec->error_list_count, 0);
+ init_completion(&qp->qp_sec->error_complete);
+ ret = security_ib_alloc_security(&qp->qp_sec->security);
+ if (ret)
+ kfree(qp->qp_sec);
+
+ return ret;
+}
+EXPORT_SYMBOL(ib_create_qp_security);
+
+void ib_destroy_qp_security_begin(struct ib_qp_security *sec)
+{
+ mutex_lock(&sec->mutex);
+
+ /* Remove the QP from the lists so it won't get added to
+ * a to_error_list during the destroy process.
+ */
+ if (sec->ports_pkeys) {
+ port_pkey_list_remove(&sec->ports_pkeys->main);
+ port_pkey_list_remove(&sec->ports_pkeys->alt);
+ }
+
+ /* If the QP is already in one or more of those lists
+ * the destroying flag will ensure the to error flow
+ * doesn't operate on an undefined QP.
+ */
+ sec->destroying = true;
+
+ /* Record the error list count to know how many completions
+ * to wait for.
+ */
+ sec->error_comps_pending = atomic_read(&sec->error_list_count);
+
+ mutex_unlock(&sec->mutex);
+}
+
+void ib_destroy_qp_security_abort(struct ib_qp_security *sec)
+{
+ int ret;
+ int i;
+
+ /* If a concurrent cache update is in progress this
+ * QP security could be marked for an error state
+ * transition. Wait for this to complete.
+ */
+ for (i = 0; i < sec->error_comps_pending; i++)
+ wait_for_completion(&sec->error_complete);
+
+ mutex_lock(&sec->mutex);
+ sec->destroying = false;
+
+ /* Restore the position in the lists and verify
+ * access is still allowed in case a cache update
+ * occurred while attempting to destroy.
+ *
+ * Because these setting were listed already
+ * and removed during ib_destroy_qp_security_begin
+ * we know the pkey_index_qp_list for the PKey
+ * already exists so port_pkey_list_insert won't fail.
+ */
+ if (sec->ports_pkeys) {
+ port_pkey_list_insert(&sec->ports_pkeys->main);
+ port_pkey_list_insert(&sec->ports_pkeys->alt);
+ }
+
+ ret = check_qp_port_pkey_settings(sec->ports_pkeys, sec);
+ if (ret)
+ qp_to_error(sec);
+
+ mutex_unlock(&sec->mutex);
+}
+
+void ib_destroy_qp_security_end(struct ib_qp_security *sec)
+{
+ int i;
+
+ /* If a concurrent cache update is occurring we must
+ * wait until this QP security structure is processed
+ * in the QP to error flow before destroying it because
+ * the to_error_list is in use.
+ */
+ for (i = 0; i < sec->error_comps_pending; i++)
+ wait_for_completion(&sec->error_complete);
+
+ destroy_qp_security(sec);
+}
+
+void ib_security_cache_change(struct ib_device *device,
+ u8 port_num,
+ u64 subnet_prefix)
+{
+ struct pkey_index_qp_list *pkey;
+
+ list_for_each_entry(pkey,
+ &device->port_pkey_list[port_num].pkey_list,
+ pkey_index_list) {
+ check_pkey_qps(pkey,
+ device,
+ port_num,
+ subnet_prefix);
+ }
+}
+
+void ib_security_destroy_port_pkey_list(struct ib_device *device)
+{
+ struct pkey_index_qp_list *pkey, *tmp_pkey;
+ int i;
+
+ for (i = rdma_start_port(device); i <= rdma_end_port(device); i++) {
+ spin_lock(&device->port_pkey_list[i].list_lock);
+ list_for_each_entry_safe(pkey,
+ tmp_pkey,
+ &device->port_pkey_list[i].pkey_list,
+ pkey_index_list) {
+ list_del(&pkey->pkey_index_list);
+ kfree(pkey);
+ }
+ spin_unlock(&device->port_pkey_list[i].list_lock);
+ }
+}
+
+int ib_security_modify_qp(struct ib_qp *qp,
+ struct ib_qp_attr *qp_attr,
+ int qp_attr_mask,
+ struct ib_udata *udata)
+{
+ int ret = 0;
+ struct ib_ports_pkeys *tmp_pps;
+ struct ib_ports_pkeys *new_pps;
+ bool special_qp = (qp->qp_type == IB_QPT_SMI ||
+ qp->qp_type == IB_QPT_GSI);
+ bool pps_change = ((qp_attr_mask & (IB_QP_PKEY_INDEX | IB_QP_PORT)) ||
+ (qp_attr_mask & IB_QP_ALT_PATH));
+
+ if (pps_change && !special_qp) {
+ mutex_lock(&qp->qp_sec->mutex);
+ new_pps = get_new_pps(qp,
+ qp_attr,
+ qp_attr_mask);
+
+ /* Add this QP to the lists for the new port
+ * and pkey settings before checking for permission
+ * in case there is a concurrent cache update
+ * occurring. Walking the list for a cache change
+ * doesn't acquire the security mutex unless it's
+ * sending the QP to error.
+ */
+ ret = port_pkey_list_insert(&new_pps->main);
+
+ if (!ret)
+ ret = port_pkey_list_insert(&new_pps->alt);
+
+ if (!ret)
+ ret = check_qp_port_pkey_settings(new_pps,
+ qp->qp_sec);
+ }
+
+ if (!ret)
+ ret = qp->device->modify_qp(qp->real_qp,
+ qp_attr,
+ qp_attr_mask,
+ udata);
+
+ if (pps_change && !special_qp) {
+ /* Clean up the lists and free the appropriate
+ * ports_pkeys structure.
+ */
+ if (ret) {
+ tmp_pps = new_pps;
+ } else {
+ tmp_pps = qp->qp_sec->ports_pkeys;
+ qp->qp_sec->ports_pkeys = new_pps;
+ }
+
+ if (tmp_pps) {
+ port_pkey_list_remove(&tmp_pps->main);
+ port_pkey_list_remove(&tmp_pps->alt);
+ }
+ kfree(tmp_pps);
+ mutex_unlock(&qp->qp_sec->mutex);
+ }
+ return ret;
+}
+EXPORT_SYMBOL(ib_security_modify_qp);
+
+#endif /* CONFIG_SECURITY_INFINIBAND */
diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index cb3f515a..ceb9154 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -37,7 +37,7 @@
#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/sched.h>
-
+#include <linux/security.h>
#include <asm/uaccess.h>
#include "uverbs.h"
@@ -1915,6 +1915,10 @@ static int create_qp(struct ib_uverbs_file *file,
}
if (cmd->qp_type != IB_QPT_XRC_TGT) {
+ ret = ib_create_qp_security(qp, device);
+ if (ret)
+ goto err_destroy;
+
qp->real_qp = qp;
qp->device = device;
qp->pd = pd;
@@ -2405,10 +2409,18 @@ ssize_t ib_uverbs_modify_qp(struct ib_uverbs_file *file,
ret = ib_resolve_eth_dmac(qp, attr, &cmd.attr_mask);
if (ret)
goto release_qp;
- ret = qp->device->modify_qp(qp, attr,
- modify_qp_mask(qp->qp_type, cmd.attr_mask), &udata);
+
+ ret = ib_security_modify_qp(qp,
+ attr,
+ modify_qp_mask(qp->qp_type,
+ cmd.attr_mask),
+ &udata);
} else {
- ret = ib_modify_qp(qp, attr, modify_qp_mask(qp->qp_type, cmd.attr_mask));
+ ret = ib_security_modify_qp(qp,
+ attr,
+ modify_qp_mask(qp->qp_type,
+ cmd.attr_mask),
+ NULL);
}
if (ret)
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index 8368764..9845e37 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -44,6 +44,7 @@
#include <linux/in.h>
#include <linux/in6.h>
#include <net/addrconf.h>
+#include <linux/security.h>
#include <rdma/ib_verbs.h>
#include <rdma/ib_cache.h>
@@ -708,12 +709,20 @@ static struct ib_qp *__ib_open_qp(struct ib_qp *real_qp,
{
struct ib_qp *qp;
unsigned long flags;
+ int err;
qp = kzalloc(sizeof *qp, GFP_KERNEL);
if (!qp)
return ERR_PTR(-ENOMEM);
qp->real_qp = real_qp;
+ err = ib_open_shared_qp_security(qp, real_qp->device);
+ if (err) {
+ kfree(qp);
+ return ERR_PTR(err);
+ }
+
+ qp->real_qp = real_qp;
atomic_inc(&real_qp->usecnt);
qp->device = real_qp->device;
qp->event_handler = event_handler;
@@ -799,6 +808,12 @@ struct ib_qp *ib_create_qp(struct ib_pd *pd,
if (IS_ERR(qp))
return qp;
+ ret = ib_create_qp_security(qp, device);
+ if (ret) {
+ ib_destroy_qp(qp);
+ return ERR_PTR(ret);
+ }
+
qp->device = device;
qp->real_qp = qp;
qp->uobject = NULL;
@@ -1257,7 +1272,7 @@ int ib_modify_qp(struct ib_qp *qp,
if (ret)
return ret;
- return qp->device->modify_qp(qp->real_qp, qp_attr, qp_attr_mask, NULL);
+ return ib_security_modify_qp(qp->real_qp, qp_attr, qp_attr_mask, NULL);
}
EXPORT_SYMBOL(ib_modify_qp);
@@ -1286,6 +1301,7 @@ int ib_close_qp(struct ib_qp *qp)
spin_unlock_irqrestore(&real_qp->device->event_handler_lock, flags);
atomic_dec(&real_qp->usecnt);
+ ib_close_shared_qp_security(qp->qp_sec);
kfree(qp);
return 0;
@@ -1326,6 +1342,7 @@ int ib_destroy_qp(struct ib_qp *qp)
struct ib_cq *scq, *rcq;
struct ib_srq *srq;
struct ib_rwq_ind_table *ind_tbl;
+ struct ib_qp_security *sec;
int ret;
WARN_ON_ONCE(qp->mrs_used > 0);
@@ -1341,6 +1358,9 @@ int ib_destroy_qp(struct ib_qp *qp)
rcq = qp->recv_cq;
srq = qp->srq;
ind_tbl = qp->rwq_ind_tbl;
+ sec = qp->qp_sec;
+ if (sec)
+ ib_destroy_qp_security_begin(sec);
if (!qp->uobject)
rdma_rw_cleanup_mrs(qp);
@@ -1357,6 +1377,11 @@ int ib_destroy_qp(struct ib_qp *qp)
atomic_dec(&srq->usecnt);
if (ind_tbl)
atomic_dec(&ind_tbl->usecnt);
+ if (sec)
+ ib_destroy_qp_security_end(sec);
+ } else {
+ if (sec)
+ ib_destroy_qp_security_abort(sec);
}
return ret;
diff --git a/include/linux/lsm_hooks.h b/include/linux/lsm_hooks.h
index 558adfa..d576efc 100644
--- a/include/linux/lsm_hooks.h
+++ b/include/linux/lsm_hooks.h
@@ -8,6 +8,7 @@
* Copyright (C) 2001 Silicon Graphics, Inc. (Trust Technology Group)
* Copyright (C) 2015 Intel Corporation.
* Copyright (C) 2015 Casey Schaufler <casey@schaufler-ca.com>
+ * Copyright (C) 2016 Mellanox Techonologies
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -903,6 +904,21 @@
* associated with the TUN device's security structure.
* @security pointer to the TUN devices's security structure.
*
+ * Security hooks for Infiniband
+ *
+ * @ib_pkey_access:
+ * Check permission to access a pkey when modifing a QP.
+ * @subnet_prefix the subnet prefix of the port being used.
+ * @pkey the pkey to be accessed.
+ * @sec pointer to a security structure.
+ * @ib_alloc_security:
+ * Allocate a security structure for Infiniband objects.
+ * @sec pointer to a security structure pointer.
+ * Returns 0 on success, non-zero on failure
+ * @ib_free_security:
+ * Deallocate an Infiniband security structure.
+ * @sec contains the security structure to be freed.
+ *
* Security hooks for XFRM operations.
*
* @xfrm_policy_alloc_security:
@@ -1611,6 +1627,12 @@
int (*tun_dev_open)(void *security);
#endif /* CONFIG_SECURITY_NETWORK */
+#ifdef CONFIG_SECURITY_INFINIBAND
+ int (*ib_pkey_access)(void *sec, u64 subnet_prefix, u16 pkey);
+ int (*ib_alloc_security)(void **sec);
+ void (*ib_free_security)(void *sec);
+#endif /* CONFIG_SECURITY_INFINIBAND */
+
#ifdef CONFIG_SECURITY_NETWORK_XFRM
int (*xfrm_policy_alloc_security)(struct xfrm_sec_ctx **ctxp,
struct xfrm_user_sec_ctx *sec_ctx,
@@ -1841,6 +1863,11 @@ struct security_hook_heads {
struct list_head tun_dev_attach;
struct list_head tun_dev_open;
#endif /* CONFIG_SECURITY_NETWORK */
+#ifdef CONFIG_SECURITY_INFINIBAND
+ struct list_head ib_pkey_access;
+ struct list_head ib_alloc_security;
+ struct list_head ib_free_security;
+#endif /* CONFIG_SECURITY_INFINIBAND */
#ifdef CONFIG_SECURITY_NETWORK_XFRM
struct list_head xfrm_policy_alloc_security;
struct list_head xfrm_policy_clone_security;
diff --git a/include/linux/security.h b/include/linux/security.h
index c2125e9..342ca4c 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -6,6 +6,7 @@
* Copyright (C) 2001 Networks Associates Technology, Inc <ssmalley@nai.com>
* Copyright (C) 2001 James Morris <jmorris@intercode.com.au>
* Copyright (C) 2001 Silicon Graphics, Inc. (Trust Technology Group)
+ * Copyright (C) 2016 Mellanox Techonologies
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -1393,6 +1394,26 @@ static inline int security_tun_dev_open(void *security)
}
#endif /* CONFIG_SECURITY_NETWORK */
+#ifdef CONFIG_SECURITY_INFINIBAND
+int security_ib_pkey_access(void *sec, u64 subnet_prefix, u16 pkey);
+int security_ib_alloc_security(void **sec);
+void security_ib_free_security(void *sec);
+#else /* CONFIG_SECURITY_INFINIBAND */
+static inline int security_ib_pkey_access(void *sec, u64 subnet_prefix, u16 pkey)
+{
+ return 0;
+}
+
+static inline int security_ib_alloc_security(void **sec)
+{
+ return 0;
+}
+
+static inline void security_ib_free_security(void *sec)
+{
+}
+#endif /* CONFIG_SECURITY_INFINIBAND */
+
#ifdef CONFIG_SECURITY_NETWORK_XFRM
int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp,
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index db178fd..1143586 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -1511,6 +1511,45 @@ struct ib_rwq_ind_table_init_attr {
struct ib_wq **ind_tbl;
};
+enum port_pkey_state {
+ IB_PORT_PKEY_NOT_VALID = 0,
+ IB_PORT_PKEY_VALID = 1,
+ IB_PORT_PKEY_LISTED = 2,
+};
+
+struct ib_qp_security;
+
+struct ib_port_pkey {
+ enum port_pkey_state state;
+ u16 pkey_index;
+ u8 port_num;
+ struct list_head qp_list;
+ struct list_head to_error_list;
+ struct ib_qp_security *sec;
+};
+
+struct ib_ports_pkeys {
+ struct ib_port_pkey main;
+ struct ib_port_pkey alt;
+};
+
+struct ib_qp_security {
+ struct ib_qp *qp;
+ struct ib_device *dev;
+ /* Hold this mutex when changing port and pkey settings. */
+ struct mutex mutex;
+ struct ib_ports_pkeys *ports_pkeys;
+ /* A list of all open shared QP handles. Required to enforce security
+ * properly for all users of a shared QP.
+ */
+ struct list_head shared_qp_list;
+ void *security;
+ bool destroying;
+ atomic_t error_list_count;
+ struct completion error_complete;
+ int error_comps_pending;
+};
+
/*
* @max_write_sge: Maximum SGE elements per RDMA WRITE request.
* @max_read_sge: Maximum SGE elements per RDMA READ request.
@@ -1540,6 +1579,7 @@ struct ib_qp {
u32 max_read_sge;
enum ib_qp_type qp_type;
struct ib_rwq_ind_table *rwq_ind_tbl;
+ struct ib_qp_security *qp_sec;
};
struct ib_mr {
@@ -1820,6 +1860,12 @@ struct ib_port_immutable {
u32 max_mad_size;
};
+struct ib_port_pkey_list {
+ /* Lock to hold while modifying the list. */
+ spinlock_t list_lock;
+ struct list_head pkey_list;
+};
+
struct ib_device {
struct device *dma_device;
@@ -1842,6 +1888,8 @@ struct ib_device {
int num_comp_vectors;
+ struct ib_port_pkey_list *port_pkey_list;
+
struct iw_cm_verbs *iwcm;
/**
diff --git a/security/Kconfig b/security/Kconfig
index 118f454..85cf893 100644
--- a/security/Kconfig
+++ b/security/Kconfig
@@ -49,6 +49,15 @@ config SECURITY_NETWORK
implement socket and networking access controls.
If you are unsure how to answer this question, answer N.
+config SECURITY_INFINIBAND
+ bool "Infiniband Security Hooks"
+ depends on SECURITY && INFINIBAND
+ help
+ This enables the Infiniband security hooks.
+ If enabled, a security module can use these hooks to
+ implement Infiniband access controls.
+ If you are unsure how to answer this question, answer N.
+
config SECURITY_NETWORK_XFRM
bool "XFRM (IPSec) Networking Security Hooks"
depends on XFRM && SECURITY_NETWORK
diff --git a/security/security.c b/security/security.c
index f825304..7d3bf2f 100644
--- a/security/security.c
+++ b/security/security.c
@@ -4,6 +4,7 @@
* Copyright (C) 2001 WireX Communications, Inc <chris@wirex.com>
* Copyright (C) 2001-2002 Greg Kroah-Hartman <greg@kroah.com>
* Copyright (C) 2001 Networks Associates Technology, Inc <ssmalley@nai.com>
+ * Copyright (C) 2016 Mellanox Technologies
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -1441,6 +1442,27 @@ int security_tun_dev_open(void *security)
#endif /* CONFIG_SECURITY_NETWORK */
+#ifdef CONFIG_SECURITY_INFINIBAND
+
+int security_ib_pkey_access(void *sec, u64 subnet_prefix, u16 pkey)
+{
+ return call_int_hook(ib_pkey_access, 0, sec, subnet_prefix, pkey);
+}
+EXPORT_SYMBOL(security_ib_pkey_access);
+
+int security_ib_alloc_security(void **sec)
+{
+ return call_int_hook(ib_alloc_security, 0, sec);
+}
+EXPORT_SYMBOL(security_ib_alloc_security);
+
+void security_ib_free_security(void *sec)
+{
+ call_void_hook(ib_free_security, sec);
+}
+EXPORT_SYMBOL(security_ib_free_security);
+#endif /* CONFIG_SECURITY_INFINIBAND */
+
#ifdef CONFIG_SECURITY_NETWORK_XFRM
int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp,
@@ -1898,6 +1920,15 @@ struct security_hook_heads security_hook_heads = {
LIST_HEAD_INIT(security_hook_heads.tun_dev_attach),
.tun_dev_open = LIST_HEAD_INIT(security_hook_heads.tun_dev_open),
#endif /* CONFIG_SECURITY_NETWORK */
+
+#ifdef CONFIG_SECURITY_INFINIBAND
+ .ib_pkey_access = LIST_HEAD_INIT(security_hook_heads.ib_pkey_access),
+ .ib_alloc_security =
+ LIST_HEAD_INIT(security_hook_heads.ib_alloc_security),
+ .ib_free_security =
+ LIST_HEAD_INIT(security_hook_heads.ib_free_security),
+#endif /* CONFIG_SECURITY_INFINIBAND */
+
#ifdef CONFIG_SECURITY_NETWORK_XFRM
.xfrm_policy_alloc_security =
LIST_HEAD_INIT(security_hook_heads.xfrm_policy_alloc_security),
--
1.8.3.1
^ permalink raw reply related
* [PATCH v4 3/9] selinux lsm IB/core: Implement LSM notification system
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw-69jw2NvuJkxg9hUCZPvPmw, paul-r2n+y4ga6xFZroRs9YW3xA,
sds-+05T5uksL2qpZYMLLGbcSA, eparis-FjpueFixGhCM4zKIHC2jIg,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
sean.hefty-ral2JQCrhuEAvxtiuMwx3w,
hal.rosenstock-Re5JQEeQqe8AvxtiuMwx3w
Cc: selinux-+05T5uksL2qpZYMLLGbcSA,
linux-security-module-u79uwXL29TY76Z2rM5mHXA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA,
yevgenyp-VPRAkNaXOzVWk0Htik3J/w, liranl-VPRAkNaXOzVWk0Htik3J/w,
leonro-VPRAkNaXOzVWk0Htik3J/w, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
From: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Add a generic notificaiton mechanism in the LSM. Interested consumers
can register a callback with the LSM and security modules can produce
events.
Because access to Infiniband QPs are enforced in the setup phase of a
connection security should be enforced again if the policy changes.
Register infiniband devices for policy change notification and check all
QPs on that device when the notification is received.
Add a call to the notification mechanism from SELinux when the AVC
cache changes or setenforce is cleared.
issue: 736423
Change-Id: I9f5a13026c9219105ee5e0cd99edf8eabdafe946
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
v2:
- new patch that has the generic notification, replaces selinux and
IB/core patches related to the ib_flush callback. Yuval Shaia and Paul
Moore
v3:
- use notifier chains. Paul Moore
v4:
- Seperate avc callback for LSM notifier. Paul Moore
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
drivers/infiniband/core/device.c | 53 ++++++++++++++++++++++++++++++++++++++++
include/linux/security.h | 8 ++++++
security/security.c | 20 +++++++++++++++
security/selinux/hooks.c | 11 +++++++++
security/selinux/selinuxfs.c | 2 ++
5 files changed, 94 insertions(+)
diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index 5b42e83..7b6fd06 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -39,6 +39,8 @@
#include <linux/init.h>
#include <linux/mutex.h>
#include <linux/netdevice.h>
+#include <linux/security.h>
+#include <linux/notifier.h>
#include <rdma/rdma_netlink.h>
#include <rdma/ib_addr.h>
#include <rdma/ib_cache.h>
@@ -82,6 +84,14 @@ struct ib_client_data {
static DEFINE_MUTEX(device_mutex);
static DECLARE_RWSEM(lists_rwsem);
+static int ib_security_change(struct notifier_block *nb, unsigned long event,
+ void *lsm_data);
+static void ib_policy_change_task(struct work_struct *work);
+static DECLARE_WORK(ib_policy_change_work, ib_policy_change_task);
+
+static struct notifier_block ibdev_lsm_nb = {
+ .notifier_call = ib_security_change,
+};
static int ib_device_check_mandatory(struct ib_device *device)
{
@@ -344,6 +354,40 @@ static int setup_port_pkey_list(struct ib_device *device)
return 0;
}
+static void ib_policy_change_task(struct work_struct *work)
+{
+ struct ib_device *dev;
+
+ down_read(&lists_rwsem);
+ list_for_each_entry(dev, &device_list, core_list) {
+ int i;
+
+ for (i = rdma_start_port(dev); i <= rdma_end_port(dev); i++) {
+ u64 sp;
+ int ret = ib_get_cached_subnet_prefix(dev,
+ i,
+ &sp);
+
+ WARN_ONCE(ret,
+ "ib_get_cached_subnet_prefix err: %d, this should never happen here\n",
+ ret);
+ ib_security_cache_change(dev, i, sp);
+ }
+ }
+ up_read(&lists_rwsem);
+}
+
+static int ib_security_change(struct notifier_block *nb, unsigned long event,
+ void *lsm_data)
+{
+ if (event != LSM_POLICY_CHANGE)
+ return NOTIFY_DONE;
+
+ schedule_work(&ib_policy_change_work);
+
+ return NOTIFY_OK;
+}
+
/**
* ib_register_device - Register an IB device with IB core
* @device:Device to register
@@ -1075,10 +1119,18 @@ static int __init ib_core_init(void)
goto err_sa;
}
+ ret = register_lsm_notifier(&ibdev_lsm_nb);
+ if (ret) {
+ pr_warn("Couldn't register LSM notifier. ret %d\n", ret);
+ goto err_ibnl_clients;
+ }
+
ib_cache_setup();
return 0;
+err_ibnl_clients:
+ ib_remove_ibnl_clients();
err_sa:
ib_sa_cleanup();
err_mad:
@@ -1098,6 +1150,7 @@ static int __init ib_core_init(void)
static void __exit ib_core_cleanup(void)
{
+ unregister_lsm_notifier(&ibdev_lsm_nb);
ib_cache_cleanup();
ib_remove_ibnl_clients();
ib_sa_cleanup();
diff --git a/include/linux/security.h b/include/linux/security.h
index 342ca4c..d8c29cd 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -69,6 +69,14 @@
struct user_namespace;
struct timezone;
+enum lsm_event {
+ LSM_POLICY_CHANGE,
+};
+
+int call_lsm_notifier(enum lsm_event event, void *data);
+int register_lsm_notifier(struct notifier_block *nb);
+int unregister_lsm_notifier(struct notifier_block *nb);
+
/* These functions are in security/commoncap.c */
extern int cap_capable(const struct cred *cred, struct user_namespace *ns,
int cap, int audit);
diff --git a/security/security.c b/security/security.c
index 7d3bf2f..40326d4 100644
--- a/security/security.c
+++ b/security/security.c
@@ -33,6 +33,8 @@
/* Maximum number of letters for an LSM name string */
#define SECURITY_NAME_MAX 10
+static ATOMIC_NOTIFIER_HEAD(lsm_notifier_chain);
+
/* Boot-time LSM user choice */
static __initdata char chosen_lsm[SECURITY_NAME_MAX + 1] =
CONFIG_DEFAULT_SECURITY;
@@ -98,6 +100,24 @@ int __init security_module_enable(const char *module)
return !strcmp(module, chosen_lsm);
}
+int call_lsm_notifier(enum lsm_event event, void *data)
+{
+ return atomic_notifier_call_chain(&lsm_notifier_chain, event, data);
+}
+EXPORT_SYMBOL(call_lsm_notifier);
+
+int register_lsm_notifier(struct notifier_block *nb)
+{
+ return atomic_notifier_chain_register(&lsm_notifier_chain, nb);
+}
+EXPORT_SYMBOL(register_lsm_notifier);
+
+int unregister_lsm_notifier(struct notifier_block *nb)
+{
+ return atomic_notifier_chain_unregister(&lsm_notifier_chain, nb);
+}
+EXPORT_SYMBOL(unregister_lsm_notifier);
+
/*
* Hook list operation macros.
*
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 09fd610..7578198 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -170,6 +170,14 @@ static int selinux_netcache_avc_callback(u32 event)
return 0;
}
+static int selinux_lsm_notifier_avc_callback(u32 event)
+{
+ if (event == AVC_CALLBACK_RESET) {
+ call_lsm_notifier(LSM_POLICY_CHANGE, NULL);
+ }
+ return 0;
+}
+
/*
* initialise the security for the init task
*/
@@ -6325,6 +6333,9 @@ static __init int selinux_init(void)
if (avc_add_callback(selinux_netcache_avc_callback, AVC_CALLBACK_RESET))
panic("SELinux: Unable to register AVC netcache callback\n");
+ if (avc_add_callback(selinux_lsm_notifier_avc_callback, AVC_CALLBACK_RESET))
+ panic("SELinux: Unable to register AVC LSM notifier callback\n");
+
if (selinux_enforcing)
printk(KERN_DEBUG "SELinux: Starting in enforcing mode\n");
else
diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c
index 72c145d..d3f9192 100644
--- a/security/selinux/selinuxfs.c
+++ b/security/selinux/selinuxfs.c
@@ -177,6 +177,8 @@ static ssize_t sel_write_enforce(struct file *file, const char __user *buf,
avc_ss_reset(0);
selnl_notify_setenforce(selinux_enforcing);
selinux_status_update_setenforce(selinux_enforcing);
+ if (!selinux_enforcing)
+ call_lsm_notifier(LSM_POLICY_CHANGE, NULL);
}
length = count;
out:
--
1.8.3.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH v4 4/9] IB/core: Enforce security on management datagrams
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw-69jw2NvuJkxg9hUCZPvPmw, paul-r2n+y4ga6xFZroRs9YW3xA,
sds-+05T5uksL2qpZYMLLGbcSA, eparis-FjpueFixGhCM4zKIHC2jIg,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
sean.hefty-ral2JQCrhuEAvxtiuMwx3w,
hal.rosenstock-Re5JQEeQqe8AvxtiuMwx3w
Cc: selinux-+05T5uksL2qpZYMLLGbcSA,
linux-security-module-u79uwXL29TY76Z2rM5mHXA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA,
yevgenyp-VPRAkNaXOzVWk0Htik3J/w, liranl-VPRAkNaXOzVWk0Htik3J/w,
leonro-VPRAkNaXOzVWk0Htik3J/w, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
From: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Allocate and free a security context when creating and destroying a MAD
agent. This context is used for controlling access to PKeys and sending
and receiving SMPs.
When sending or receiving a MAD check that the agent has permission to
access the PKey for the Subnet Prefix of the port.
During MAD and snoop agent registration for SMI QPs check that the
calling process has permission to access the manage the subnet and
register a callback with the LSM to be notified of policy changes. When
notificaiton of a policy change occurs recheck permission and set a flag
indicating sending and receiving SMPs is allowed.
When sending and receiving MADs check that the agent has access to the
SMI if it's on an SMI QP. Because security policy can change it's
possible permission was allowed when creating the agent, but no longer
is.
issue: 736423
Change-Id: I17343224da7ffffe7ee0cc67dadd44c473e453d0
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
v2:
- Squashed LSM hook additions. Paul Moore
- Changed security blobs to void*. Paul Moore
- Shorten end_port to port. Paul Moore
- Change "smp" to "manage_subnet". Paul Moore
- Use the LSM policy change notification and a flag to track permission
instead of calling the LSM hook for every SMP. Dan Jurgens
- Squashed PKey and SMP enforcement into the same patch and moved the
logic into security.c. Dan Jurgens
v3:
- ib_port -> ib_endport. Paul Moore
- Use notifier chains for LSM notification. Paul Moore
- Reorder LSM hook parameters to put sec first. Paul Moore
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
drivers/infiniband/core/core_priv.h | 35 ++++++++++++++
drivers/infiniband/core/mad.c | 52 +++++++++++++++++----
drivers/infiniband/core/security.c | 92 +++++++++++++++++++++++++++++++++++++
include/linux/lsm_hooks.h | 8 ++++
include/linux/security.h | 6 +++
include/rdma/ib_mad.h | 4 ++
security/security.c | 8 ++++
7 files changed, 197 insertions(+), 8 deletions(-)
diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h
index 68e3de0..2c35162 100644
--- a/drivers/infiniband/core/core_priv.h
+++ b/drivers/infiniband/core/core_priv.h
@@ -37,6 +37,8 @@
#include <linux/spinlock.h>
#include <rdma/ib_verbs.h>
+#include <rdma/ib_mad.h>
+#include "mad_priv.h"
struct pkey_index_qp_list {
struct list_head pkey_index_list;
@@ -166,6 +168,11 @@ int ib_get_cached_subnet_prefix(struct ib_device *device,
u64 *sn_pfx);
#ifdef CONFIG_SECURITY_INFINIBAND
+int ib_security_pkey_access(struct ib_device *dev,
+ u8 port_num,
+ u16 pkey_index,
+ void *sec);
+
void ib_security_destroy_port_pkey_list(struct ib_device *device);
void ib_security_cache_change(struct ib_device *device,
@@ -183,7 +190,19 @@ int ib_security_modify_qp(struct ib_qp *qp,
void ib_destroy_qp_security_end(struct ib_qp_security *sec);
int ib_open_shared_qp_security(struct ib_qp *qp, struct ib_device *dev);
void ib_close_shared_qp_security(struct ib_qp_security *sec);
+int ib_mad_agent_security_setup(struct ib_mad_agent *agent,
+ enum ib_qp_type qp_type);
+void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent);
+int ib_mad_enforce_security(struct ib_mad_agent_private *map, u16 pkey_index);
#else
+static inline int ib_security_pkey_access(struct ib_device *dev,
+ u8 port_num,
+ u16 pkey_index,
+ void *sec)
+{
+ return 0;
+}
+
static inline void ib_security_destroy_port_pkey_list(struct ib_device *device)
{
}
@@ -232,5 +251,21 @@ static inline int ib_open_shared_qp_security(struct ib_qp *qp,
static inline void ib_close_shared_qp_security(struct ib_qp_security *sec)
{
}
+
+static inline int ib_mad_agent_security_setup(struct ib_mad_agent *agent,
+ enum ib_qp_type qp_type)
+{
+ return 0;
+}
+
+static inline void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent)
+{
+}
+
+static inline int ib_mad_enforce_security(struct ib_mad_agent_private *map,
+ u16 pkey_index)
+{
+ return 0;
+}
#endif
#endif /* _CORE_PRIV_H */
diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
index 40cbd6b..b6041ec 100644
--- a/drivers/infiniband/core/mad.c
+++ b/drivers/infiniband/core/mad.c
@@ -40,9 +40,11 @@
#include <linux/dma-mapping.h>
#include <linux/slab.h>
#include <linux/module.h>
+#include <linux/security.h>
#include <rdma/ib_cache.h>
#include "mad_priv.h"
+#include "core_priv.h"
#include "mad_rmpp.h"
#include "smi.h"
#include "opa_smi.h"
@@ -367,6 +369,12 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device,
atomic_set(&mad_agent_priv->refcount, 1);
init_completion(&mad_agent_priv->comp);
+ ret2 = ib_mad_agent_security_setup(&mad_agent_priv->agent, qp_type);
+ if (ret2) {
+ ret = ERR_PTR(ret2);
+ goto error4;
+ }
+
spin_lock_irqsave(&port_priv->reg_lock, flags);
mad_agent_priv->agent.hi_tid = ++ib_mad_client_id;
@@ -384,7 +392,7 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device,
if (method) {
if (method_in_use(&method,
mad_reg_req))
- goto error4;
+ goto error5;
}
}
ret2 = add_nonoui_reg_req(mad_reg_req, mad_agent_priv,
@@ -400,14 +408,14 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device,
if (is_vendor_method_in_use(
vendor_class,
mad_reg_req))
- goto error4;
+ goto error5;
}
}
ret2 = add_oui_reg_req(mad_reg_req, mad_agent_priv);
}
if (ret2) {
ret = ERR_PTR(ret2);
- goto error4;
+ goto error5;
}
}
@@ -416,9 +424,10 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device,
spin_unlock_irqrestore(&port_priv->reg_lock, flags);
return &mad_agent_priv->agent;
-
-error4:
+error5:
spin_unlock_irqrestore(&port_priv->reg_lock, flags);
+ ib_mad_agent_security_cleanup(&mad_agent_priv->agent);
+error4:
kfree(reg_req);
error3:
kfree(mad_agent_priv);
@@ -489,6 +498,7 @@ struct ib_mad_agent *ib_register_mad_snoop(struct ib_device *device,
struct ib_mad_agent *ret;
struct ib_mad_snoop_private *mad_snoop_priv;
int qpn;
+ int err;
/* Validate parameters */
if ((is_snooping_sends(mad_snoop_flags) && !snoop_handler) ||
@@ -523,17 +533,25 @@ struct ib_mad_agent *ib_register_mad_snoop(struct ib_device *device,
mad_snoop_priv->agent.port_num = port_num;
mad_snoop_priv->mad_snoop_flags = mad_snoop_flags;
init_completion(&mad_snoop_priv->comp);
+
+ err = ib_mad_agent_security_setup(&mad_snoop_priv->agent, qp_type);
+ if (err) {
+ ret = ERR_PTR(err);
+ goto error2;
+ }
+
mad_snoop_priv->snoop_index = register_snoop_agent(
&port_priv->qp_info[qpn],
mad_snoop_priv);
if (mad_snoop_priv->snoop_index < 0) {
ret = ERR_PTR(mad_snoop_priv->snoop_index);
- goto error2;
+ goto error3;
}
atomic_set(&mad_snoop_priv->refcount, 1);
return &mad_snoop_priv->agent;
-
+error3:
+ ib_mad_agent_security_cleanup(&mad_snoop_priv->agent);
error2:
kfree(mad_snoop_priv);
error1:
@@ -579,6 +597,8 @@ static void unregister_mad_agent(struct ib_mad_agent_private *mad_agent_priv)
deref_mad_agent(mad_agent_priv);
wait_for_completion(&mad_agent_priv->comp);
+ ib_mad_agent_security_cleanup(&mad_agent_priv->agent);
+
kfree(mad_agent_priv->reg_req);
kfree(mad_agent_priv);
}
@@ -597,6 +617,8 @@ static void unregister_mad_snoop(struct ib_mad_snoop_private *mad_snoop_priv)
deref_snoop_agent(mad_snoop_priv);
wait_for_completion(&mad_snoop_priv->comp);
+ ib_mad_agent_security_cleanup(&mad_snoop_priv->agent);
+
kfree(mad_snoop_priv);
}
@@ -1219,12 +1241,16 @@ int ib_post_send_mad(struct ib_mad_send_buf *send_buf,
/* Walk list of send WRs and post each on send list */
for (; send_buf; send_buf = next_send_buf) {
-
mad_send_wr = container_of(send_buf,
struct ib_mad_send_wr_private,
send_buf);
mad_agent_priv = mad_send_wr->mad_agent_priv;
+ ret = ib_mad_enforce_security(mad_agent_priv,
+ mad_send_wr->send_wr.pkey_index);
+ if (ret)
+ goto error;
+
if (!send_buf->mad_agent->send_handler ||
(send_buf->timeout_ms &&
!send_buf->mad_agent->recv_handler)) {
@@ -1958,6 +1984,14 @@ static void ib_mad_complete_recv(struct ib_mad_agent_private *mad_agent_priv,
struct ib_mad_send_wr_private *mad_send_wr;
struct ib_mad_send_wc mad_send_wc;
unsigned long flags;
+ int ret;
+
+ ret = ib_mad_enforce_security(mad_agent_priv,
+ mad_recv_wc->wc->pkey_index);
+ if (ret) {
+ ib_free_recv_mad(mad_recv_wc);
+ deref_mad_agent(mad_agent_priv);
+ }
INIT_LIST_HEAD(&mad_recv_wc->rmpp_list);
list_add(&mad_recv_wc->recv_buf.list, &mad_recv_wc->rmpp_list);
@@ -2015,6 +2049,8 @@ static void ib_mad_complete_recv(struct ib_mad_agent_private *mad_agent_priv,
mad_recv_wc);
deref_mad_agent(mad_agent_priv);
}
+
+ return;
}
static enum smi_action handle_ib_smi(const struct ib_mad_port_private *port_priv,
diff --git a/drivers/infiniband/core/security.c b/drivers/infiniband/core/security.c
index 45800dd..bc09af5 100644
--- a/drivers/infiniband/core/security.c
+++ b/drivers/infiniband/core/security.c
@@ -39,6 +39,7 @@
#include <rdma/ib_verbs.h>
#include <rdma/ib_cache.h>
#include "core_priv.h"
+#include "mad_priv.h"
static struct pkey_index_qp_list *get_pkey_idx_qp_list(struct ib_port_pkey *pp)
{
@@ -614,4 +615,95 @@ int ib_security_modify_qp(struct ib_qp *qp,
}
EXPORT_SYMBOL(ib_security_modify_qp);
+int ib_security_pkey_access(struct ib_device *dev,
+ u8 port_num,
+ u16 pkey_index,
+ void *sec)
+{
+ u64 subnet_prefix;
+ u16 pkey;
+ int ret;
+
+ ret = ib_get_cached_pkey(dev, port_num, pkey_index, &pkey);
+ if (ret)
+ return ret;
+
+ ret = ib_get_cached_subnet_prefix(dev, port_num, &subnet_prefix);
+
+ if (ret)
+ return ret;
+
+ return security_ib_pkey_access(sec, subnet_prefix, pkey);
+}
+EXPORT_SYMBOL(ib_security_pkey_access);
+
+static int ib_mad_agent_security_change(struct notifier_block *nb,
+ unsigned long event,
+ void *data)
+{
+ struct ib_mad_agent *ag = container_of(nb, struct ib_mad_agent, lsm_nb);
+
+ if (event != LSM_POLICY_CHANGE)
+ return NOTIFY_DONE;
+
+ ag->smp_allowed = !security_ib_endport_manage_subnet(ag->security,
+ ag->device->name,
+ ag->port_num);
+
+ return NOTIFY_OK;
+}
+
+int ib_mad_agent_security_setup(struct ib_mad_agent *agent,
+ enum ib_qp_type qp_type)
+{
+ int ret;
+
+ ret = security_ib_alloc_security(&agent->security);
+ if (ret)
+ return ret;
+
+ if (qp_type != IB_QPT_SMI)
+ return 0;
+
+ ret = security_ib_endport_manage_subnet(agent->security,
+ agent->device->name,
+ agent->port_num);
+ if (ret)
+ return ret;
+
+ agent->lsm_nb.notifier_call = ib_mad_agent_security_change;
+ ret = register_lsm_notifier(&agent->lsm_nb);
+ if (ret)
+ return ret;
+
+ agent->smp_allowed = true;
+ agent->lsm_nb_reg = true;
+ return 0;
+}
+
+void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent)
+{
+ security_ib_free_security(agent->security);
+ if (agent->lsm_nb_reg)
+ unregister_lsm_notifier(&agent->lsm_nb);
+}
+
+int ib_mad_enforce_security(struct ib_mad_agent_private *map, u16 pkey_index)
+{
+ int ret;
+
+ if (map->agent.qp->qp_type == IB_QPT_SMI && !map->agent.smp_allowed)
+ return -EACCES;
+
+ ret = ib_security_pkey_access(map->agent.device,
+ map->agent.port_num,
+ pkey_index,
+ map->agent.security);
+
+ if (ret)
+ return ret;
+
+ return 0;
+}
+
#endif /* CONFIG_SECURITY_INFINIBAND */
diff --git a/include/linux/lsm_hooks.h b/include/linux/lsm_hooks.h
index d576efc..3936216 100644
--- a/include/linux/lsm_hooks.h
+++ b/include/linux/lsm_hooks.h
@@ -911,6 +911,11 @@
* @subnet_prefix the subnet prefix of the port being used.
* @pkey the pkey to be accessed.
* @sec pointer to a security structure.
+ * @ib_endport_manage_subnet:
+ * Check permissions to send and receive SMPs on a end port.
+ * @dev_name the IB device name (i.e. mlx4_0).
+ * @port_num the port number.
+ * @sec pointer to a security structure.
* @ib_alloc_security:
* Allocate a security structure for Infiniband objects.
* @sec pointer to a security structure pointer.
@@ -1629,6 +1634,8 @@
#ifdef CONFIG_SECURITY_INFINIBAND
int (*ib_pkey_access)(void *sec, u64 subnet_prefix, u16 pkey);
+ int (*ib_endport_manage_subnet)(void *sec, const char *dev_name,
+ u8 port_num);
int (*ib_alloc_security)(void **sec);
void (*ib_free_security)(void *sec);
#endif /* CONFIG_SECURITY_INFINIBAND */
@@ -1865,6 +1872,7 @@ struct security_hook_heads {
#endif /* CONFIG_SECURITY_NETWORK */
#ifdef CONFIG_SECURITY_INFINIBAND
struct list_head ib_pkey_access;
+ struct list_head ib_endport_manage_subnet;
struct list_head ib_alloc_security;
struct list_head ib_free_security;
#endif /* CONFIG_SECURITY_INFINIBAND */
diff --git a/include/linux/security.h b/include/linux/security.h
index d8c29cd..9241f86 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1404,6 +1404,7 @@ static inline int security_tun_dev_open(void *security)
#ifdef CONFIG_SECURITY_INFINIBAND
int security_ib_pkey_access(void *sec, u64 subnet_prefix, u16 pkey);
+int security_ib_endport_manage_subnet(void *sec, const char *name, u8 port_num);
int security_ib_alloc_security(void **sec);
void security_ib_free_security(void *sec);
#else /* CONFIG_SECURITY_INFINIBAND */
@@ -1412,6 +1413,11 @@ static inline int security_ib_pkey_access(void *sec, u64 subnet_prefix, u16 pkey
return 0;
}
+static inline int security_ib_endport_manage_subnet(void *sec, const char *dev_name, u8 port_num)
+{
+ return 0;
+}
+
static inline int security_ib_alloc_security(void **sec)
{
return 0;
diff --git a/include/rdma/ib_mad.h b/include/rdma/ib_mad.h
index c8a773f..23c4a4e 100644
--- a/include/rdma/ib_mad.h
+++ b/include/rdma/ib_mad.h
@@ -537,6 +537,10 @@ struct ib_mad_agent {
u32 flags;
u8 port_num;
u8 rmpp_version;
+ void *security;
+ bool smp_allowed;
+ bool lsm_nb_reg;
+ struct notifier_block lsm_nb;
};
/**
diff --git a/security/security.c b/security/security.c
index 40326d4..e4fd2b6 100644
--- a/security/security.c
+++ b/security/security.c
@@ -1470,6 +1470,12 @@ int security_ib_pkey_access(void *sec, u64 subnet_prefix, u16 pkey)
}
EXPORT_SYMBOL(security_ib_pkey_access);
+int security_ib_endport_manage_subnet(void *sec, const char *dev_name, u8 port_num)
+{
+ return call_int_hook(ib_endport_manage_subnet, 0, sec, dev_name, port_num);
+}
+EXPORT_SYMBOL(security_ib_endport_manage_subnet);
+
int security_ib_alloc_security(void **sec)
{
return call_int_hook(ib_alloc_security, 0, sec);
@@ -1943,6 +1949,8 @@ struct security_hook_heads security_hook_heads = {
#ifdef CONFIG_SECURITY_INFINIBAND
.ib_pkey_access = LIST_HEAD_INIT(security_hook_heads.ib_pkey_access),
+ .ib_endport_manage_subnet =
+ LIST_HEAD_INIT(security_hook_heads.ib_endport_manage_subnet),
.ib_alloc_security =
LIST_HEAD_INIT(security_hook_heads.ib_alloc_security),
.ib_free_security =
--
1.8.3.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH v4 5/9] selinux: Create policydb version for Infiniband support
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw, paul, sds, eparis, dledford, sean.hefty, hal.rosenstock
Cc: selinux, linux-security-module, linux-rdma, yevgenyp, liranl,
leonro, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj@mellanox.com>
From: Daniel Jurgens <danielj@mellanox.com>
Support for Infiniband requires the addition of two new object contexts,
one for infiniband PKeys and another IB Ports. Added handlers to read
and write the new ocontext types when reading or writing a binary policy
representation.
issue: 736423
Change-Id: I906bf7a1e0bd59c3cccdce278e5fe98ba8c404b8
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Reviewed-by: Eli Cohen <eli@mellanox.com>
---
v2:
- Shorten ib_end_port to ib_port. Paul Moore
- Added bounds checking to port number. Paul Moore
- Eliminated {} in OCON_PKEY case statement. Yuval Shaia
v3:
- ib_port -> ib_endport. Paul Moore
v4:
- removed unneeded brackets in ocontext_read. Paul Moore
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
---
security/selinux/include/security.h | 3 +-
security/selinux/ss/policydb.c | 129 +++++++++++++++++++++++++++++++-----
security/selinux/ss/policydb.h | 27 +++++---
3 files changed, 135 insertions(+), 24 deletions(-)
diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h
index 308a286..6bb9b0a 100644
--- a/security/selinux/include/security.h
+++ b/security/selinux/include/security.h
@@ -36,10 +36,11 @@
#define POLICYDB_VERSION_DEFAULT_TYPE 28
#define POLICYDB_VERSION_CONSTRAINT_NAMES 29
#define POLICYDB_VERSION_XPERMS_IOCTL 30
+#define POLICYDB_VERSION_INFINIBAND 31
/* Range of policy versions we understand*/
#define POLICYDB_VERSION_MIN POLICYDB_VERSION_BASE
-#define POLICYDB_VERSION_MAX POLICYDB_VERSION_XPERMS_IOCTL
+#define POLICYDB_VERSION_MAX POLICYDB_VERSION_INFINIBAND
/* Mask for just the mount related flags */
#define SE_MNTMASK 0x0f
diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c
index d719db4..24e16da 100644
--- a/security/selinux/ss/policydb.c
+++ b/security/selinux/ss/policydb.c
@@ -17,6 +17,11 @@
*
* Added support for the policy capability bitmap
*
+ * Update: Mellanox Techonologies
+ *
+ * Added Infiniband support
+ *
+ * Copyright (C) 2016 Mellanox Techonologies
* Copyright (C) 2007 Hewlett-Packard Development Company, L.P.
* Copyright (C) 2004-2005 Trusted Computer Solutions, Inc.
* Copyright (C) 2003 - 2004 Tresys Technology, LLC
@@ -76,81 +81,86 @@ struct policydb_compat_info {
{
.version = POLICYDB_VERSION_BASE,
.sym_num = SYM_NUM - 3,
- .ocon_num = OCON_NUM - 1,
+ .ocon_num = OCON_NUM - 3,
},
{
.version = POLICYDB_VERSION_BOOL,
.sym_num = SYM_NUM - 2,
- .ocon_num = OCON_NUM - 1,
+ .ocon_num = OCON_NUM - 3,
},
{
.version = POLICYDB_VERSION_IPV6,
.sym_num = SYM_NUM - 2,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_NLCLASS,
.sym_num = SYM_NUM - 2,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_MLS,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_AVTAB,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_RANGETRANS,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_POLCAP,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_PERMISSIVE,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_BOUNDARY,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_FILENAME_TRANS,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_ROLETRANS,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_NEW_OBJECT_DEFAULTS,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_DEFAULT_TYPE,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_CONSTRAINT_NAMES,
.sym_num = SYM_NUM,
- .ocon_num = OCON_NUM,
+ .ocon_num = OCON_NUM - 2,
},
{
.version = POLICYDB_VERSION_XPERMS_IOCTL,
.sym_num = SYM_NUM,
+ .ocon_num = OCON_NUM - 2,
+ },
+ {
+ .version = POLICYDB_VERSION_INFINIBAND,
+ .sym_num = SYM_NUM,
.ocon_num = OCON_NUM,
},
};
@@ -2222,6 +2232,60 @@ static int ocontext_read(struct policydb *p, struct policydb_compat_info *info,
goto out;
break;
}
+ case OCON_PKEY:
+ rc = next_entry(nodebuf, fp, sizeof(u32) * 6);
+ if (rc)
+ goto out;
+
+ c->u.pkey.subnet_prefix = be64_to_cpu(*((__be64 *)nodebuf));
+ /* The subnet prefix is stored as an IPv6
+ * address in the policy.
+ *
+ * Check that the lower 2 DWORDS are 0.
+ */
+ if (nodebuf[2] || nodebuf[3]) {
+ rc = -EINVAL;
+ goto out;
+ }
+
+ if (nodebuf[4] > 0xffff ||
+ nodebuf[5] > 0xffff) {
+ rc = -EINVAL;
+ goto out;
+ }
+
+ c->u.pkey.low_pkey = le32_to_cpu(nodebuf[4]);
+ c->u.pkey.high_pkey = le32_to_cpu(nodebuf[5]);
+
+ rc = context_read_and_validate(&c->context[0],
+ p,
+ fp);
+ if (rc)
+ goto out;
+ break;
+ case OCON_IB_ENDPORT:
+ rc = next_entry(buf, fp, sizeof(u32) * 2);
+ if (rc)
+ goto out;
+ len = le32_to_cpu(buf[0]);
+
+ rc = str_read(&c->u.ib_endport.dev_name, GFP_KERNEL, fp, len);
+ if (rc)
+ goto out;
+
+ if (buf[1] > 0xff || buf[1] == 0) {
+ rc = -EINVAL;
+ goto out;
+ }
+
+ c->u.ib_endport.port_num = le32_to_cpu(buf[1]);
+
+ rc = context_read_and_validate(&c->context[0],
+ p,
+ fp);
+ if (rc)
+ goto out;
+ break;
}
}
}
@@ -3151,6 +3215,41 @@ static int ocontext_write(struct policydb *p, struct policydb_compat_info *info,
if (rc)
return rc;
break;
+ case OCON_PKEY:
+ *((__be64 *)nodebuf) = cpu_to_be64(c->u.pkey.subnet_prefix);
+
+ /*
+ * The low order 2 bits were confirmed to be 0
+ * when the policy was loaded. Write them out
+ * as zero
+ */
+ nodebuf[2] = 0;
+ nodebuf[3] = 0;
+
+ nodebuf[4] = cpu_to_le32(c->u.pkey.low_pkey);
+ nodebuf[5] = cpu_to_le32(c->u.pkey.high_pkey);
+
+ rc = put_entry(nodebuf, sizeof(u32), 6, fp);
+ if (rc)
+ return rc;
+ rc = context_write(p, &c->context[0], fp);
+ if (rc)
+ return rc;
+ break;
+ case OCON_IB_ENDPORT:
+ len = strlen(c->u.ib_endport.dev_name);
+ buf[0] = cpu_to_le32(len);
+ buf[1] = cpu_to_le32(c->u.ib_endport.port_num);
+ rc = put_entry(buf, sizeof(u32), 2, fp);
+ if (rc)
+ return rc;
+ rc = put_entry(c->u.ib_endport.dev_name, 1, len, fp);
+ if (rc)
+ return rc;
+ rc = context_write(p, &c->context[0], fp);
+ if (rc)
+ return rc;
+ break;
}
}
}
diff --git a/security/selinux/ss/policydb.h b/security/selinux/ss/policydb.h
index 725d594..edb329d 100644
--- a/security/selinux/ss/policydb.h
+++ b/security/selinux/ss/policydb.h
@@ -187,6 +187,15 @@ struct ocontext {
u32 addr[4];
u32 mask[4];
} node6; /* IPv6 node information */
+ struct {
+ u64 subnet_prefix;
+ u16 low_pkey;
+ u16 high_pkey;
+ } pkey;
+ struct {
+ char *dev_name;
+ u8 port_num;
+ } ib_endport;
} u;
union {
u32 sclass; /* security class for genfs */
@@ -215,14 +224,16 @@ struct genfs {
#define SYM_NUM 8
/* object context array indices */
-#define OCON_ISID 0 /* initial SIDs */
-#define OCON_FS 1 /* unlabeled file systems */
-#define OCON_PORT 2 /* TCP and UDP port numbers */
-#define OCON_NETIF 3 /* network interfaces */
-#define OCON_NODE 4 /* nodes */
-#define OCON_FSUSE 5 /* fs_use */
-#define OCON_NODE6 6 /* IPv6 nodes */
-#define OCON_NUM 7
+#define OCON_ISID 0 /* initial SIDs */
+#define OCON_FS 1 /* unlabeled file systems */
+#define OCON_PORT 2 /* TCP and UDP port numbers */
+#define OCON_NETIF 3 /* network interfaces */
+#define OCON_NODE 4 /* nodes */
+#define OCON_FSUSE 5 /* fs_use */
+#define OCON_NODE6 6 /* IPv6 nodes */
+#define OCON_PKEY 7 /* Infiniband PKeys */
+#define OCON_IB_ENDPORT 8 /* Infiniband end ports */
+#define OCON_NUM 9
/* The policy database */
struct policydb {
--
1.8.3.1
^ permalink raw reply related
* [PATCH v4 6/9] selinux: Allocate and free infiniband security hooks
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw, paul, sds, eparis, dledford, sean.hefty, hal.rosenstock
Cc: selinux, linux-security-module, linux-rdma, yevgenyp, liranl,
leonro, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj@mellanox.com>
From: Daniel Jurgens <danielj@mellanox.com>
Implement and attach hooks to allocate and free Infiniband object
security structures.
issue: 736423
Change-Id: I3bdbecee7aab6d7615a02967c39a5a8792a14d44
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
---
v2:
- Use void * blobs for security structs. Paul Moore
- Shorten ib_end_port to ib_port. Paul Moore
- Allocate memory for security struct with GFP_KERNEL. Yuval Shaia
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
---
security/selinux/hooks.c | 25 ++++++++++++++++++++++++-
security/selinux/include/objsec.h | 5 +++++
2 files changed, 29 insertions(+), 1 deletion(-)
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 7578198..f48759d 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -17,6 +17,7 @@
* Paul Moore <paul@paul-moore.com>
* Copyright (C) 2007 Hitachi Software Engineering Co., Ltd.
* Yuichi Nakamura <ynakam@hitachisoft.jp>
+ * Copyright (C) 2016 Mellanox Technologies
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2,
@@ -6082,7 +6083,26 @@ static int selinux_key_getsecurity(struct key *key, char **_buffer)
*_buffer = context;
return rc;
}
+#endif
+
+#ifdef CONFIG_SECURITY_INFINIBAND
+static int selinux_ib_alloc_security(void **ib_sec)
+{
+ struct ib_security_struct *sec;
+
+ sec = kzalloc(sizeof(*sec), GFP_KERNEL);
+ if (!sec)
+ return -ENOMEM;
+ sec->sid = current_sid();
+
+ *ib_sec = sec;
+ return 0;
+}
+static void selinux_ib_free_security(void *ib_sec)
+{
+ kfree(ib_sec);
+}
#endif
static struct security_hook_list selinux_hooks[] = {
@@ -6269,7 +6289,10 @@ static int selinux_key_getsecurity(struct key *key, char **_buffer)
LSM_HOOK_INIT(tun_dev_attach_queue, selinux_tun_dev_attach_queue),
LSM_HOOK_INIT(tun_dev_attach, selinux_tun_dev_attach),
LSM_HOOK_INIT(tun_dev_open, selinux_tun_dev_open),
-
+#ifdef CONFIG_SECURITY_INFINIBAND
+ LSM_HOOK_INIT(ib_alloc_security, selinux_ib_alloc_security),
+ LSM_HOOK_INIT(ib_free_security, selinux_ib_free_security),
+#endif
#ifdef CONFIG_SECURITY_NETWORK_XFRM
LSM_HOOK_INIT(xfrm_policy_alloc_security, selinux_xfrm_policy_alloc),
LSM_HOOK_INIT(xfrm_policy_clone_security, selinux_xfrm_policy_clone),
diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h
index c21e135..8e7db43 100644
--- a/security/selinux/include/objsec.h
+++ b/security/selinux/include/objsec.h
@@ -10,6 +10,7 @@
*
* Copyright (C) 2001,2002 Networks Associates Technology, Inc.
* Copyright (C) 2003 Red Hat, Inc., James Morris <jmorris@redhat.com>
+ * Copyright (C) 2016 Mellanox Technologies
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2,
@@ -128,6 +129,10 @@ struct key_security_struct {
u32 sid; /* SID of key */
};
+struct ib_security_struct {
+ u32 sid; /* SID of the queue pair or MAD agent */
+};
+
extern unsigned int selinux_checkreqprot;
#endif /* _SELINUX_OBJSEC_H_ */
--
1.8.3.1
^ permalink raw reply related
* [PATCH v4 7/9] selinux: Implement Infiniband PKey "Access" access vector
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw-69jw2NvuJkxg9hUCZPvPmw, paul-r2n+y4ga6xFZroRs9YW3xA,
sds-+05T5uksL2qpZYMLLGbcSA, eparis-FjpueFixGhCM4zKIHC2jIg,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
sean.hefty-ral2JQCrhuEAvxtiuMwx3w,
hal.rosenstock-Re5JQEeQqe8AvxtiuMwx3w
Cc: selinux-+05T5uksL2qpZYMLLGbcSA,
linux-security-module-u79uwXL29TY76Z2rM5mHXA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA,
yevgenyp-VPRAkNaXOzVWk0Htik3J/w, liranl-VPRAkNaXOzVWk0Htik3J/w,
leonro-VPRAkNaXOzVWk0Htik3J/w, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
From: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Add a type and access vector for PKeys. Implement the ib_pkey_access
hook to check that the caller has permission to access the PKey on the
given subnet prefix. Add an interface to get the PKey SID. Walk the PKey
ocontexts to find an entry for the given subnet prefix and pkey.
issue: 736423
Change-Id: Ib3f7df9b5f98da77a6048855d146f9dcaceba336
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
v2:
- Use void* blobs for security structs. Paul Moore
- Add pkey specific data to the audit log. Paul Moore
- Don't introduce a new initial sid, use unlabeled. Stephen Smalley
v3:
- Reorder parameters to pkey_access hook. Paul Moore
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
include/linux/lsm_audit.h | 7 +++++++
security/lsm_audit.c | 13 ++++++++++++
security/selinux/hooks.c | 23 +++++++++++++++++++++
security/selinux/include/classmap.h | 2 ++
security/selinux/include/security.h | 2 ++
security/selinux/ss/services.c | 41 +++++++++++++++++++++++++++++++++++++
6 files changed, 88 insertions(+)
diff --git a/include/linux/lsm_audit.h b/include/linux/lsm_audit.h
index e58e577..402b770 100644
--- a/include/linux/lsm_audit.h
+++ b/include/linux/lsm_audit.h
@@ -45,6 +45,11 @@ struct lsm_ioctlop_audit {
u16 cmd;
};
+struct lsm_pkey_audit {
+ u64 subnet_prefix;
+ u16 pkey;
+};
+
/* Auxiliary data to use in generating the audit record. */
struct common_audit_data {
char type;
@@ -60,6 +65,7 @@ struct common_audit_data {
#define LSM_AUDIT_DATA_DENTRY 10
#define LSM_AUDIT_DATA_IOCTL_OP 11
#define LSM_AUDIT_DATA_FILE 12
+#define LSM_AUDIT_DATA_PKEY 13
union {
struct path path;
struct dentry *dentry;
@@ -77,6 +83,7 @@ struct common_audit_data {
char *kmod_name;
struct lsm_ioctlop_audit *op;
struct file *file;
+ struct lsm_pkey_audit *pkey;
} u;
/* this union contains LSM specific data */
union {
diff --git a/security/lsm_audit.c b/security/lsm_audit.c
index 37f04da..b18d277 100644
--- a/security/lsm_audit.c
+++ b/security/lsm_audit.c
@@ -410,6 +410,19 @@ static void dump_common_audit_data(struct audit_buffer *ab,
audit_log_format(ab, " kmod=");
audit_log_untrustedstring(ab, a->u.kmod_name);
break;
+ case LSM_AUDIT_DATA_PKEY: {
+ struct in6_addr sbn_pfx;
+
+ memset(&sbn_pfx.s6_addr, 0,
+ sizeof(sbn_pfx.s6_addr));
+
+ memcpy(&sbn_pfx.s6_addr, &a->u.pkey->subnet_prefix,
+ sizeof(a->u.pkey->subnet_prefix));
+
+ audit_log_format(ab, " pkey=0x%x subnet_prefix=%pI6c",
+ a->u.pkey->pkey, &sbn_pfx);
+ break;
+ }
} /* switch (a->type) */
}
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index f48759d..20fb292 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -6086,6 +6086,28 @@ static int selinux_key_getsecurity(struct key *key, char **_buffer)
#endif
#ifdef CONFIG_SECURITY_INFINIBAND
+static int selinux_ib_pkey_access(void *ib_sec, u64 subnet_prefix, u16 pkey_val)
+{
+ struct common_audit_data ad;
+ int err;
+ u32 sid = 0;
+ struct ib_security_struct *sec = ib_sec;
+ struct lsm_pkey_audit pkey;
+
+ err = security_pkey_sid(subnet_prefix, pkey_val, &sid);
+
+ if (err)
+ return err;
+
+ ad.type = LSM_AUDIT_DATA_PKEY;
+ pkey.subnet_prefix = subnet_prefix;
+ pkey.pkey = pkey_val;
+ ad.u.pkey = &pkey;
+ return avc_has_perm(sec->sid, sid,
+ SECCLASS_INFINIBAND_PKEY,
+ INFINIBAND_PKEY__ACCESS, &ad);
+}
+
static int selinux_ib_alloc_security(void **ib_sec)
{
struct ib_security_struct *sec;
@@ -6290,6 +6312,7 @@ static void selinux_ib_free_security(void *ib_sec)
LSM_HOOK_INIT(tun_dev_attach, selinux_tun_dev_attach),
LSM_HOOK_INIT(tun_dev_open, selinux_tun_dev_open),
#ifdef CONFIG_SECURITY_INFINIBAND
+ LSM_HOOK_INIT(ib_pkey_access, selinux_ib_pkey_access),
LSM_HOOK_INIT(ib_alloc_security, selinux_ib_alloc_security),
LSM_HOOK_INIT(ib_free_security, selinux_ib_free_security),
#endif
diff --git a/security/selinux/include/classmap.h b/security/selinux/include/classmap.h
index 1f1f4b2..d42dd4d 100644
--- a/security/selinux/include/classmap.h
+++ b/security/selinux/include/classmap.h
@@ -165,5 +165,7 @@ struct security_class_mapping secclass_map[] = {
{ COMMON_CAP_PERMS, NULL } },
{ "cap2_userns",
{ COMMON_CAP2_PERMS, NULL } },
+ { "infiniband_pkey",
+ { "access", NULL } },
{ NULL }
};
diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h
index 6bb9b0a..17afb7c 100644
--- a/security/selinux/include/security.h
+++ b/security/selinux/include/security.h
@@ -176,6 +176,8 @@ int security_get_user_sids(u32 callsid, char *username,
int security_port_sid(u8 protocol, u16 port, u32 *out_sid);
+int security_pkey_sid(u64 subnet_prefix, u16 pkey_num, u32 *out_sid);
+
int security_netif_sid(char *name, u32 *if_sid);
int security_node_sid(u16 domain, void *addr, u32 addrlen,
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 082b20c..085c54b 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2203,6 +2203,47 @@ int security_port_sid(u8 protocol, u16 port, u32 *out_sid)
}
/**
+ * security_pkey_sid - Obtain the SID for a pkey.
+ * @subnet_prefix: Subnet Prefix
+ * @pkey_num: pkey number
+ * @out_sid: security identifier
+ */
+int security_pkey_sid(u64 subnet_prefix, u16 pkey_num, u32 *out_sid)
+{
+ struct ocontext *c;
+ int rc = 0;
+
+ read_lock(&policy_rwlock);
+
+ c = policydb.ocontexts[OCON_PKEY];
+ while (c) {
+ if (c->u.pkey.low_pkey <= pkey_num &&
+ c->u.pkey.high_pkey >= pkey_num &&
+ c->u.pkey.subnet_prefix == subnet_prefix)
+ break;
+
+ c = c->next;
+ }
+
+ if (c) {
+ if (!c->sid[0]) {
+ rc = sidtab_context_to_sid(&sidtab,
+ &c->context[0],
+ &c->sid[0]);
+ if (rc)
+ goto out;
+ }
+ *out_sid = c->sid[0];
+ } else {
+ *out_sid = SECINITSID_UNLABELED;
+ }
+
+out:
+ read_unlock(&policy_rwlock);
+ return rc;
+}
+
+/**
* security_netif_sid - Obtain the SID for a network interface.
* @name: interface name
* @if_sid: interface SID
--
1.8.3.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH v4 8/9] selinux: Add IB Port SMP access vector
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw-69jw2NvuJkxg9hUCZPvPmw, paul-r2n+y4ga6xFZroRs9YW3xA,
sds-+05T5uksL2qpZYMLLGbcSA, eparis-FjpueFixGhCM4zKIHC2jIg,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
sean.hefty-ral2JQCrhuEAvxtiuMwx3w,
hal.rosenstock-Re5JQEeQqe8AvxtiuMwx3w
Cc: selinux-+05T5uksL2qpZYMLLGbcSA,
linux-security-module-u79uwXL29TY76Z2rM5mHXA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA,
yevgenyp-VPRAkNaXOzVWk0Htik3J/w, liranl-VPRAkNaXOzVWk0Htik3J/w,
leonro-VPRAkNaXOzVWk0Htik3J/w, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
From: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Add a type for Infiniband ports and an access vector for subnet
management packets. Implement the ib_port_smp hook to check that the
caller has permission to send and receive SMPs on the end port specified
by the device name and port. Add interface to query the SID for a IB
port, which walks the IB_PORT ocontexts to find an entry for the
given name and port.
issue: 736423
Change-Id: If8b365f3cf32e77a2060073f1a53e27ea846804d
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
v2:
- Shorted ib_end_port. Paul Moore
- Pass void blobs to security hooks. Paul Moore
- Log specific IB port info in audit log. Paul Moore
- Don't create a new intial sid, use unlabeled. Stephen Smalley
- Changed "smp" to "manage_subnet". Paul Moore
v3:
- ib_port -> ib_endport. Paul Moore
- Don't log device name as untrusted string. Paul Moore
- Reorder parameters of LSM hook. Paul Moore
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
include/linux/lsm_audit.h | 8 +++++++
security/lsm_audit.c | 5 +++++
security/selinux/hooks.c | 25 ++++++++++++++++++++++
security/selinux/include/classmap.h | 2 ++
security/selinux/include/security.h | 2 ++
security/selinux/ss/services.c | 42 +++++++++++++++++++++++++++++++++++++
6 files changed, 84 insertions(+)
diff --git a/include/linux/lsm_audit.h b/include/linux/lsm_audit.h
index 402b770..7047b4c 100644
--- a/include/linux/lsm_audit.h
+++ b/include/linux/lsm_audit.h
@@ -21,6 +21,7 @@
#include <linux/path.h>
#include <linux/key.h>
#include <linux/skbuff.h>
+#include <rdma/ib_verbs.h>
struct lsm_network_audit {
int netif;
@@ -50,6 +51,11 @@ struct lsm_pkey_audit {
u16 pkey;
};
+struct lsm_ib_endport_audit {
+ char dev_name[IB_DEVICE_NAME_MAX];
+ u8 port_num;
+};
+
/* Auxiliary data to use in generating the audit record. */
struct common_audit_data {
char type;
@@ -66,6 +72,7 @@ struct common_audit_data {
#define LSM_AUDIT_DATA_IOCTL_OP 11
#define LSM_AUDIT_DATA_FILE 12
#define LSM_AUDIT_DATA_PKEY 13
+#define LSM_AUDIT_DATA_IB_ENDPORT 14
union {
struct path path;
struct dentry *dentry;
@@ -84,6 +91,7 @@ struct common_audit_data {
struct lsm_ioctlop_audit *op;
struct file *file;
struct lsm_pkey_audit *pkey;
+ struct lsm_ib_endport_audit *ib_endport;
} u;
/* this union contains LSM specific data */
union {
diff --git a/security/lsm_audit.c b/security/lsm_audit.c
index b18d277..549fe9d 100644
--- a/security/lsm_audit.c
+++ b/security/lsm_audit.c
@@ -423,6 +423,11 @@ static void dump_common_audit_data(struct audit_buffer *ab,
a->u.pkey->pkey, &sbn_pfx);
break;
}
+ case LSM_AUDIT_DATA_IB_ENDPORT:
+ audit_log_format(ab, " device=%s port_num=%u",
+ a->u.ib_endport->dev_name,
+ a->u.ib_endport->port_num);
+ break;
} /* switch (a->type) */
}
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 20fb292..ea3f6d0 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -6108,6 +6108,29 @@ static int selinux_ib_pkey_access(void *ib_sec, u64 subnet_prefix, u16 pkey_val)
INFINIBAND_PKEY__ACCESS, &ad);
}
+static int selinux_ib_endport_manage_subnet(void *ib_sec, const char *dev_name,
+ u8 port_num)
+{
+ struct common_audit_data ad;
+ int err;
+ u32 sid = 0;
+ struct ib_security_struct *sec = ib_sec;
+ struct lsm_ib_endport_audit ib_endport;
+
+ err = security_ib_endport_sid(dev_name, port_num, &sid);
+
+ if (err)
+ return err;
+
+ ad.type = LSM_AUDIT_DATA_IB_ENDPORT;
+ strncpy(ib_endport.dev_name, dev_name, sizeof(ib_endport.dev_name));
+ ib_endport.port_num = port_num;
+ ad.u.ib_endport = &ib_endport;
+ return avc_has_perm(sec->sid, sid,
+ SECCLASS_INFINIBAND_ENDPORT,
+ INFINIBAND_ENDPORT__MANAGE_SUBNET, &ad);
+}
+
static int selinux_ib_alloc_security(void **ib_sec)
{
struct ib_security_struct *sec;
@@ -6313,6 +6336,8 @@ static void selinux_ib_free_security(void *ib_sec)
LSM_HOOK_INIT(tun_dev_open, selinux_tun_dev_open),
#ifdef CONFIG_SECURITY_INFINIBAND
LSM_HOOK_INIT(ib_pkey_access, selinux_ib_pkey_access),
+ LSM_HOOK_INIT(ib_endport_manage_subnet,
+ selinux_ib_endport_manage_subnet),
LSM_HOOK_INIT(ib_alloc_security, selinux_ib_alloc_security),
LSM_HOOK_INIT(ib_free_security, selinux_ib_free_security),
#endif
diff --git a/security/selinux/include/classmap.h b/security/selinux/include/classmap.h
index d42dd4d..f93b64b 100644
--- a/security/selinux/include/classmap.h
+++ b/security/selinux/include/classmap.h
@@ -167,5 +167,7 @@ struct security_class_mapping secclass_map[] = {
{ COMMON_CAP2_PERMS, NULL } },
{ "infiniband_pkey",
{ "access", NULL } },
+ { "infiniband_endport",
+ { "manage_subnet", NULL } },
{ NULL }
};
diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h
index 17afb7c..8a6e5e7 100644
--- a/security/selinux/include/security.h
+++ b/security/selinux/include/security.h
@@ -178,6 +178,8 @@ int security_get_user_sids(u32 callsid, char *username,
int security_pkey_sid(u64 subnet_prefix, u16 pkey_num, u32 *out_sid);
+int security_ib_endport_sid(const char *dev_name, u8 port_num, u32 *out_sid);
+
int security_netif_sid(char *name, u32 *if_sid);
int security_node_sid(u16 domain, void *addr, u32 addrlen,
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 085c54b..db40ce7 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2244,6 +2244,48 @@ int security_pkey_sid(u64 subnet_prefix, u16 pkey_num, u32 *out_sid)
}
/**
+ * security_ib_endport_sid - Obtain the SID for a subnet management interface.
+ * @dev_name: device name
+ * @port: port number
+ * @out_sid: security identifier
+ */
+int security_ib_endport_sid(const char *dev_name, u8 port_num, u32 *out_sid)
+{
+ struct ocontext *c;
+ int rc = 0;
+
+ read_lock(&policy_rwlock);
+
+ c = policydb.ocontexts[OCON_IB_ENDPORT];
+ while (c) {
+ if (c->u.ib_endport.port_num == port_num &&
+ !strncmp(c->u.ib_endport.dev_name,
+ dev_name,
+ IB_DEVICE_NAME_MAX))
+ break;
+
+ c = c->next;
+ }
+
+ if (c) {
+ if (!c->sid[0]) {
+ rc = sidtab_context_to_sid(&sidtab,
+ &c->context[0],
+ &c->sid[0]);
+ if (rc)
+ goto out;
+ }
+ *out_sid = c->sid[0];
+ } else {
+ *out_sid = SECINITSID_UNLABELED;
+ }
+
+out:
+ read_unlock(&policy_rwlock);
+ return rc;
+}
+
+/**
* security_netif_sid - Obtain the SID for a network interface.
* @name: interface name
* @if_sid: interface SID
--
1.8.3.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH v4 9/9] selinux: Add a cache for quicker retreival of PKey SIDs
From: Dan Jurgens @ 2016-11-08 21:06 UTC (permalink / raw)
To: chrisw-69jw2NvuJkxg9hUCZPvPmw, paul-r2n+y4ga6xFZroRs9YW3xA,
sds-+05T5uksL2qpZYMLLGbcSA, eparis-FjpueFixGhCM4zKIHC2jIg,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
sean.hefty-ral2JQCrhuEAvxtiuMwx3w,
hal.rosenstock-Re5JQEeQqe8AvxtiuMwx3w
Cc: selinux-+05T5uksL2qpZYMLLGbcSA,
linux-security-module-u79uwXL29TY76Z2rM5mHXA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA,
yevgenyp-VPRAkNaXOzVWk0Htik3J/w, liranl-VPRAkNaXOzVWk0Htik3J/w,
leonro-VPRAkNaXOzVWk0Htik3J/w, Daniel Jurgens
In-Reply-To: <1478639185-47521-1-git-send-email-danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
From: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
It is likely that the SID for the same PKey will be requested many
times. To reduce the time to modify QPs and process MADs use a cache to
store PKey SIDs.
This code is heavily based on the "netif" and "netport" concept
originally developed by James Morris <jmorris-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> and Paul Moore
<paul-r2n+y4ga6xFZroRs9YW3xA@public.gmane.org> (see security/selinux/netif.c and
security/selinux/netport.c for more information)
issue: 736423
Change-Id: I176c3079d5d84d06839b4f750100ac47a6081e94
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
v2:
- Renamed the files to ibpkey. Paul Moore
- Fixed a braket indentation mismatch in sel_pkey_find. Yuval Shaia
- Change spin_lock_bh to spin_lock_irqsave to resolve HARDIRQ lockdep
warning. Dan Jurgens
Signed-off-by: Daniel Jurgens <danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
---
security/selinux/Makefile | 2 +-
security/selinux/hooks.c | 4 +-
security/selinux/ibpkey.c | 245 ++++++++++++++++++++++++++++++++++++++
security/selinux/include/ibpkey.h | 31 +++++
security/selinux/include/objsec.h | 6 +
5 files changed, 286 insertions(+), 2 deletions(-)
create mode 100644 security/selinux/ibpkey.c
create mode 100644 security/selinux/include/ibpkey.h
diff --git a/security/selinux/Makefile b/security/selinux/Makefile
index 3411c33..ff5895e 100644
--- a/security/selinux/Makefile
+++ b/security/selinux/Makefile
@@ -5,7 +5,7 @@
obj-$(CONFIG_SECURITY_SELINUX) := selinux.o
selinux-y := avc.o hooks.o selinuxfs.o netlink.o nlmsgtab.o netif.o \
- netnode.o netport.o exports.o \
+ netnode.o netport.o ibpkey.o exports.o \
ss/ebitmap.o ss/hashtab.o ss/symtab.o ss/sidtab.o ss/avtab.o \
ss/policydb.o ss/services.o ss/conditional.o ss/mls.o ss/status.o
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index ea3f6d0..619d2ea 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -90,6 +90,7 @@
#include "netif.h"
#include "netnode.h"
#include "netport.h"
+#include "ibpkey.h"
#include "xfrm.h"
#include "netlabel.h"
#include "audit.h"
@@ -174,6 +175,7 @@ static int selinux_netcache_avc_callback(u32 event)
static int selinux_lsm_notifier_avc_callback(u32 event)
{
if (event == AVC_CALLBACK_RESET) {
+ sel_pkey_flush();
call_lsm_notifier(LSM_POLICY_CHANGE, NULL);
}
return 0;
@@ -6094,7 +6096,7 @@ static int selinux_ib_pkey_access(void *ib_sec, u64 subnet_prefix, u16 pkey_val)
struct ib_security_struct *sec = ib_sec;
struct lsm_pkey_audit pkey;
- err = security_pkey_sid(subnet_prefix, pkey_val, &sid);
+ err = sel_pkey_sid(subnet_prefix, pkey_val, &sid);
if (err)
return err;
diff --git a/security/selinux/ibpkey.c b/security/selinux/ibpkey.c
new file mode 100644
index 0000000..6e52c54
--- /dev/null
+++ b/security/selinux/ibpkey.c
@@ -0,0 +1,245 @@
+/*
+ * Pkey table
+ *
+ * SELinux must keep a mapping of Infinband PKEYs to labels/SIDs. This
+ * mapping is maintained as part of the normal policy but a fast cache is
+ * needed to reduce the lookup overhead.
+ *
+ * This code is heavily based on the "netif" and "netport" concept originally
+ * developed by
+ * James Morris <jmorris-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> and
+ * Paul Moore <paul-r2n+y4ga6xFZroRs9YW3xA@public.gmane.org>
+ * (see security/selinux/netif.c and security/selinux/netport.c for more
+ * information)
+ *
+ */
+
+/*
+ * (c) Mellanox Technologies, 2016
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+#include "ibpkey.h"
+#include "objsec.h"
+
+#define SEL_PKEY_HASH_SIZE 256
+#define SEL_PKEY_HASH_BKT_LIMIT 16
+
+struct sel_pkey_bkt {
+ int size;
+ struct list_head list;
+};
+
+struct sel_pkey {
+ struct pkey_security_struct psec;
+ struct list_head list;
+ struct rcu_head rcu;
+};
+
+static LIST_HEAD(sel_pkey_list);
+static DEFINE_SPINLOCK(sel_pkey_lock);
+static struct sel_pkey_bkt sel_pkey_hash[SEL_PKEY_HASH_SIZE];
+
+/**
+ * sel_pkey_hashfn - Hashing function for the pkey table
+ * @pkey: pkey number
+ *
+ * Description:
+ * This is the hashing function for the pkey table, it returns the bucket
+ * number for the given pkey.
+ *
+ */
+static unsigned int sel_pkey_hashfn(u16 pkey)
+{
+ return (pkey & (SEL_PKEY_HASH_SIZE - 1));
+}
+
+/**
+ * sel_pkey_find - Search for a pkey record
+ * @subnet_prefix: subnet_prefix
+ * @pkey_num: pkey_num
+ *
+ * Description:
+ * Search the pkey table and return the matching record. If an entry
+ * can not be found in the table return NULL.
+ *
+ */
+static struct sel_pkey *sel_pkey_find(u64 subnet_prefix, u16 pkey_num)
+{
+ unsigned int idx;
+ struct sel_pkey *pkey;
+
+ idx = sel_pkey_hashfn(pkey_num);
+ list_for_each_entry_rcu(pkey, &sel_pkey_hash[idx].list, list) {
+ if (pkey->psec.pkey == pkey_num &&
+ pkey->psec.subnet_prefix == subnet_prefix)
+ return pkey;
+ }
+
+ return NULL;
+}
+
+/**
+ * sel_pkey_insert - Insert a new pkey into the table
+ * @pkey: the new pkey record
+ *
+ * Description:
+ * Add a new pkey record to the hash table.
+ *
+ */
+static void sel_pkey_insert(struct sel_pkey *pkey)
+{
+ unsigned int idx;
+
+ /* we need to impose a limit on the growth of the hash table so check
+ * this bucket to make sure it is within the specified bounds
+ */
+ idx = sel_pkey_hashfn(pkey->psec.pkey);
+ list_add_rcu(&pkey->list, &sel_pkey_hash[idx].list);
+ if (sel_pkey_hash[idx].size == SEL_PKEY_HASH_BKT_LIMIT) {
+ struct sel_pkey *tail;
+
+ tail = list_entry(
+ rcu_dereference_protected(
+ sel_pkey_hash[idx].list.prev,
+ lockdep_is_held(&sel_pkey_lock)),
+ struct sel_pkey, list);
+ list_del_rcu(&tail->list);
+ kfree_rcu(tail, rcu);
+ } else {
+ sel_pkey_hash[idx].size++;
+ }
+}
+
+/**
+ * sel_pkey_sid_slow - Lookup the SID of a pkey using the policy
+ * @subnet_prefix: subnet prefix
+ * @pkey_num: pkey number
+ * @sid: pkey SID
+ *
+ * Description:
+ * This function determines the SID of a pkey by querying the security
+ * policy. The result is added to the pkey table to speedup future
+ * queries. Returns zero on success, negative values on failure.
+ *
+ */
+static int sel_pkey_sid_slow(u64 subnet_prefix, u16 pkey_num, u32 *sid)
+{
+ int ret = -ENOMEM;
+ struct sel_pkey *pkey;
+ struct sel_pkey *new = NULL;
+ unsigned long flags;
+
+ spin_lock_irqsave(&sel_pkey_lock, flags);
+ pkey = sel_pkey_find(subnet_prefix, pkey_num);
+ if (pkey) {
+ *sid = pkey->psec.sid;
+ spin_unlock_irqrestore(&sel_pkey_lock, flags);
+ return 0;
+ }
+
+ ret = security_pkey_sid(subnet_prefix, pkey_num, sid);
+ if (ret != 0)
+ goto out;
+
+ new = kzalloc(sizeof(*new), GFP_ATOMIC);
+ if (!new)
+ goto out;
+
+ new->psec.subnet_prefix = subnet_prefix;
+ new->psec.pkey = pkey_num;
+ new->psec.sid = *sid;
+ sel_pkey_insert(new);
+
+out:
+ spin_unlock_irqrestore(&sel_pkey_lock, flags);
+ if (unlikely(ret))
+ kfree(new);
+
+ return ret;
+}
+
+/**
+ * sel_pkey_sid - Lookup the SID of a PKEY
+ * @subnet_prefix: subnet_prefix
+ * @pkey_num: pkey number
+ * @sid: pkey SID
+ *
+ * Description:
+ * This function determines the SID of a PKEY using the fastest method
+ * possible. First the pkey table is queried, but if an entry can't be found
+ * then the policy is queried and the result is added to the table to speedup
+ * future queries. Returns zero on success, negative values on failure.
+ *
+ */
+int sel_pkey_sid(u64 subnet_prefix, u16 pkey_num, u32 *sid)
+{
+ struct sel_pkey *pkey;
+
+ rcu_read_lock();
+ pkey = sel_pkey_find(subnet_prefix, pkey_num);
+ if (pkey) {
+ *sid = pkey->psec.sid;
+ rcu_read_unlock();
+ return 0;
+ }
+ rcu_read_unlock();
+
+ return sel_pkey_sid_slow(subnet_prefix, pkey_num, sid);
+}
+
+/**
+ * sel_pkey_flush - Flush the entire pkey table
+ *
+ * Description:
+ * Remove all entries from the pkey table
+ *
+ */
+void sel_pkey_flush(void)
+{
+ unsigned int idx;
+ struct sel_pkey *pkey, *pkey_tmp;
+ unsigned long flags;
+
+ spin_lock_irqsave(&sel_pkey_lock, flags);
+ for (idx = 0; idx < SEL_PKEY_HASH_SIZE; idx++) {
+ list_for_each_entry_safe(pkey, pkey_tmp,
+ &sel_pkey_hash[idx].list, list) {
+ list_del_rcu(&pkey->list);
+ kfree_rcu(pkey, rcu);
+ }
+ sel_pkey_hash[idx].size = 0;
+ }
+ spin_unlock_irqrestore(&sel_pkey_lock, flags);
+}
+
+static __init int sel_pkey_init(void)
+{
+ int iter;
+
+ if (!selinux_enabled)
+ return 0;
+
+ for (iter = 0; iter < SEL_PKEY_HASH_SIZE; iter++) {
+ INIT_LIST_HEAD(&sel_pkey_hash[iter].list);
+ sel_pkey_hash[iter].size = 0;
+ }
+
+ return 0;
+}
+
+subsys_initcall(sel_pkey_init);
diff --git a/security/selinux/include/ibpkey.h b/security/selinux/include/ibpkey.h
new file mode 100644
index 0000000..387885a
--- /dev/null
+++ b/security/selinux/include/ibpkey.h
@@ -0,0 +1,31 @@
+/*
+ * pkey table
+ *
+ * SELinux must keep a mapping of pkeys to labels/SIDs. This
+ * mapping is maintained as part of the normal policy but a fast cache is
+ * needed to reduce the lookup overhead.
+ *
+ */
+
+/*
+ * (c) Mellanox Technologies, 2016
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _SELINUX_IB_PKEY_H
+#define _SELINUX_IB_PKEY_H
+
+void sel_pkey_flush(void);
+
+int sel_pkey_sid(u64 subnet_prefix, u16 pkey, u32 *sid);
+
+#endif
diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h
index 8e7db43..4139f28 100644
--- a/security/selinux/include/objsec.h
+++ b/security/selinux/include/objsec.h
@@ -133,6 +133,12 @@ struct ib_security_struct {
u32 sid; /* SID of the queue pair or MAD agent */
};
+struct pkey_security_struct {
+ u64 subnet_prefix; /* Port subnet prefix */
+ u16 pkey; /* PKey number */
+ u32 sid; /* SID of pkey */
+};
+
extern unsigned int selinux_checkreqprot;
#endif /* _SELINUX_OBJSEC_H_ */
--
1.8.3.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* Re: [PATCH v4 3/9] selinux lsm IB/core: Implement LSM notification system
From: kbuild test robot @ 2016-11-08 22:35 UTC (permalink / raw)
Cc: kbuild-all, chrisw, paul, sds, eparis, dledford, sean.hefty,
hal.rosenstock, selinux, linux-security-module, linux-rdma,
yevgenyp, liranl, leonro, Daniel Jurgens
In-Reply-To: <1478639185-47521-4-git-send-email-danielj@mellanox.com>
[-- Attachment #1: Type: text/plain, Size: 975 bytes --]
Hi Daniel,
[auto build test ERROR on rdma/master]
[also build test ERROR on v4.9-rc4]
[cannot apply to next-20161108]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
url: https://github.com/0day-ci/linux/commits/Dan-Jurgens/SELinux-support-for-Infiniband-RDMA/20161109-053432
base: https://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma.git master
config: i386-randconfig-s1-201645 (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
All errors (new ones prefixed by >>):
>> ERROR: "unregister_lsm_notifier" [drivers/infiniband/core/ib_core.ko] undefined!
>> ERROR: "register_lsm_notifier" [drivers/infiniband/core/ib_core.ko] undefined!
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 27488 bytes --]
^ permalink raw reply
* [PATCH] libpvrdma: Fix up for ABI file
From: Adit Ranadive @ 2016-11-08 23:06 UTC (permalink / raw)
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA; +Cc: Adit Ranadive
Reverted the ABI file to the kernel version. I added a new file that
includes this ABI file and duplicates only the required structures
for the library. This should be temporary until the UAPI is fixed
properly within libibverbs and the kernel.
Signed-off-by: Adit Ranadive <aditr-pghWNbHTmq7QT0dZR+AlfA@public.gmane.org>
---
buildlib/fixup-include/rdma-pvrdma-abi.h | 56 ++++++++++-------------
providers/pvrdma/cq.c | 10 ++---
providers/pvrdma/pvrdma-abi-fix.h | 76 ++++++++++++++++++++++++++++++++
providers/pvrdma/pvrdma.h | 2 +-
providers/pvrdma/pvrdma_main.c | 4 +-
providers/pvrdma/qp.c | 12 ++---
providers/pvrdma/verbs.c | 4 +-
7 files changed, 116 insertions(+), 48 deletions(-)
create mode 100644 providers/pvrdma/pvrdma-abi-fix.h
diff --git a/buildlib/fixup-include/rdma-pvrdma-abi.h b/buildlib/fixup-include/rdma-pvrdma-abi.h
index c7a38c5..5ca0e91 100644
--- a/buildlib/fixup-include/rdma-pvrdma-abi.h
+++ b/buildlib/fixup-include/rdma-pvrdma-abi.h
@@ -46,14 +46,14 @@
#ifndef __PVRDMA_ABI_H__
#define __PVRDMA_ABI_H__
-#include <infiniband/kern-abi.h>
+#include <linux/types.h>
-#define PVRDMA_UVERBS_ABI_VERSION 3
+#define PVRDMA_UVERBS_ABI_VERSION 3 /* ABI Version. */
#define PVRDMA_UAR_HANDLE_MASK 0x00FFFFFF /* Bottom 24 bits. */
-#define PVRDMA_UAR_QP_OFFSET 0 /* QP doorbell offset. */
+#define PVRDMA_UAR_QP_OFFSET 0 /* QP doorbell. */
#define PVRDMA_UAR_QP_SEND BIT(30) /* Send bit. */
#define PVRDMA_UAR_QP_RECV BIT(31) /* Recv bit. */
-#define PVRDMA_UAR_CQ_OFFSET 4 /* CQ doorbell offset. */
+#define PVRDMA_UAR_CQ_OFFSET 4 /* CQ doorbell. */
#define PVRDMA_UAR_CQ_ARM_SOL BIT(29) /* Arm solicited bit. */
#define PVRDMA_UAR_CQ_ARM BIT(30) /* Arm bit. */
#define PVRDMA_UAR_CQ_POLL BIT(31) /* Poll bit. */
@@ -129,55 +129,47 @@ enum pvrdma_wc_flags {
};
struct pvrdma_alloc_ucontext_resp {
- struct ibv_get_context_resp ibv_resp;
- __u32 qp_tab_size;
- __u32 reserved;
+ __u32 qp_tab_size;
+ __u32 reserved;
};
struct pvrdma_alloc_pd_resp {
- struct ibv_alloc_pd_resp ibv_resp;
- __u32 pdn;
- __u32 reserved;
+ __u32 pdn;
+ __u32 reserved;
};
struct pvrdma_create_cq {
- struct ibv_create_cq ibv_cmd;
- __u64 buf_addr;
- __u32 buf_size;
- __u32 reserved;
+ __u64 buf_addr;
+ __u32 buf_size;
+ __u32 reserved;
};
struct pvrdma_create_cq_resp {
- struct ibv_create_cq_resp ibv_resp;
- __u32 cqn;
- __u32 reserved;
+ __u32 cqn;
+ __u32 reserved;
};
struct pvrdma_resize_cq {
- struct ibv_resize_cq ibv_cmd;
- __u64 buf_addr;
- __u32 buf_size;
- __u32 reserved;
+ __u64 buf_addr;
+ __u32 buf_size;
+ __u32 reserved;
};
struct pvrdma_create_srq {
- struct ibv_create_srq ibv_cmd;
- __u64 buf_addr;
+ __u64 buf_addr;
};
struct pvrdma_create_srq_resp {
- struct ibv_create_srq_resp ibv_resp;
- __u32 srqn;
- __u32 reserved;
+ __u32 srqn;
+ __u32 reserved;
};
struct pvrdma_create_qp {
- struct ibv_create_qp ibv_cmd;
- __u64 rbuf_addr;
- __u64 sbuf_addr;
- __u32 rbuf_size;
- __u32 sbuf_size;
- __u64 qp_addr;
+ __u64 rbuf_addr;
+ __u64 sbuf_addr;
+ __u32 rbuf_size;
+ __u32 sbuf_size;
+ __u64 qp_addr;
};
/* PVRDMA masked atomic compare and swap */
diff --git a/providers/pvrdma/cq.c b/providers/pvrdma/cq.c
index bb4d1f7..0c7dcf6 100644
--- a/providers/pvrdma/cq.c
+++ b/providers/pvrdma/cq.c
@@ -214,8 +214,8 @@ struct ibv_cq *pvrdma_create_cq(struct ibv_context *context, int cqe,
int comp_vector)
{
struct pvrdma_device *dev = to_vdev(context->device);
- struct pvrdma_create_cq cmd;
- struct pvrdma_create_cq_resp resp;
+ struct user_pvrdma_create_cq cmd;
+ struct user_pvrdma_create_cq_resp resp;
struct pvrdma_cq *cq;
int ret;
@@ -239,15 +239,15 @@ struct ibv_cq *pvrdma_create_cq(struct ibv_context *context, int cqe,
cq->ring_state = cq->buf.buf;
- cmd.buf_addr = (uintptr_t) cq->buf.buf;
- cmd.buf_size = cq->buf.length;
+ cmd.udata.buf_addr = (uintptr_t) cq->buf.buf;
+ cmd.udata.buf_size = cq->buf.length;
ret = ibv_cmd_create_cq(context, cqe, channel, comp_vector,
&cq->ibv_cq, &cmd.ibv_cmd, sizeof(cmd),
&resp.ibv_resp, sizeof(resp));
if (ret)
goto err_buf;
- cq->cqn = resp.cqn;
+ cq->cqn = resp.udata.cqn;
cq->cqe_cnt = cq->ibv_cq.cqe;
return &cq->ibv_cq;
diff --git a/providers/pvrdma/pvrdma-abi-fix.h b/providers/pvrdma/pvrdma-abi-fix.h
new file mode 100644
index 0000000..3d49aca
--- /dev/null
+++ b/providers/pvrdma/pvrdma-abi-fix.h
@@ -0,0 +1,76 @@
+/*
+ * Copyright (c) 2012-2016 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of EITHER the GNU General Public License
+ * version 2 as published by the Free Software Foundation or the BSD
+ * 2-Clause License. This program is distributed in the hope that it
+ * will be useful, but WITHOUT ANY WARRANTY; WITHOUT EVEN THE IMPLIED
+ * WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+ * See the GNU General Public License version 2 for more details at
+ * http://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program available in the file COPYING in the main
+ * directory of this source tree.
+ *
+ * The BSD 2-Clause License
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+ * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
+ * COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT,
+ * INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+ * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
+ * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
+ * OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef __PVRDMA_ABI_FIX_H__
+#define __PVRDMA_ABI_FIX_H__
+
+#include <rdma/pvrdma-abi.h>
+
+struct user_pvrdma_alloc_ucontext_resp {
+ struct ibv_get_context_resp ibv_resp;
+ struct pvrdma_alloc_ucontext_resp udata;
+};
+
+struct user_pvrdma_alloc_pd_resp {
+ struct ibv_alloc_pd_resp ibv_resp;
+ struct pvrdma_alloc_pd_resp udata;
+};
+
+struct user_pvrdma_create_cq {
+ struct ibv_create_cq ibv_cmd;
+ struct pvrdma_create_cq udata;
+};
+
+struct user_pvrdma_create_cq_resp {
+ struct ibv_create_cq_resp ibv_resp;
+ struct pvrdma_create_cq_resp udata;
+};
+
+struct user_pvrdma_create_qp {
+ struct ibv_create_qp ibv_cmd;
+ struct pvrdma_create_qp udata;
+};
+
+#endif /* __PVRDMA_ABI_FIX_H__ */
diff --git a/providers/pvrdma/pvrdma.h b/providers/pvrdma/pvrdma.h
index 1c9f9f7..29c9991 100644
--- a/providers/pvrdma/pvrdma.h
+++ b/providers/pvrdma/pvrdma.h
@@ -55,11 +55,11 @@
#include <netinet/in.h>
#include <sys/mman.h>
#include <infiniband/driver.h>
-#include <rdma/pvrdma-abi.h>
#include <ccan/minmax.h>
#define BIT(nr) (1UL << (nr))
+#include "pvrdma-abi-fix.h"
#include "pvrdma_ring.h"
#ifndef likely
diff --git a/providers/pvrdma/pvrdma_main.c b/providers/pvrdma/pvrdma_main.c
index 909cf1e..9a7e07b 100644
--- a/providers/pvrdma/pvrdma_main.c
+++ b/providers/pvrdma/pvrdma_main.c
@@ -97,14 +97,14 @@ static int pvrdma_init_context_shared(struct pvrdma_context *context,
int cmd_fd)
{
struct ibv_get_context cmd;
- struct pvrdma_alloc_ucontext_resp resp;
+ struct user_pvrdma_alloc_ucontext_resp resp;
context->ibv_ctx.cmd_fd = cmd_fd;
if (ibv_cmd_get_context(&context->ibv_ctx, &cmd, sizeof(cmd),
&resp.ibv_resp, sizeof(resp)))
return errno;
- context->qp_tbl = calloc(resp.qp_tab_size & 0xFFFF,
+ context->qp_tbl = calloc(resp.udata.qp_tab_size & 0xFFFF,
sizeof(struct pvrdma_qp *));
if (!context->qp_tbl)
return -ENOMEM;
diff --git a/providers/pvrdma/qp.c b/providers/pvrdma/qp.c
index 8a37b7f..8b7c5a7 100644
--- a/providers/pvrdma/qp.c
+++ b/providers/pvrdma/qp.c
@@ -100,7 +100,7 @@ struct ibv_qp *pvrdma_create_qp(struct ibv_pd *pd,
struct ibv_qp_init_attr *attr)
{
struct pvrdma_device *dev = to_vdev(pd->context->device);
- struct pvrdma_create_qp cmd;
+ struct user_pvrdma_create_qp cmd;
struct ibv_create_qp_resp resp;
struct pvrdma_qp *qp;
int ret;
@@ -152,11 +152,11 @@ struct ibv_qp *pvrdma_create_qp(struct ibv_pd *pd,
goto err_free;
memset(&cmd, 0, sizeof(cmd));
- cmd.rbuf_addr = (uintptr_t)qp->rbuf.buf;
- cmd.rbuf_size = qp->rbuf.length;
- cmd.sbuf_addr = (uintptr_t)qp->sbuf.buf;
- cmd.sbuf_size = qp->sbuf.length;
- cmd.qp_addr = (uintptr_t) qp;
+ cmd.udata.rbuf_addr = (uintptr_t)qp->rbuf.buf;
+ cmd.udata.rbuf_size = qp->rbuf.length;
+ cmd.udata.sbuf_addr = (uintptr_t)qp->sbuf.buf;
+ cmd.udata.sbuf_size = qp->sbuf.length;
+ cmd.udata.qp_addr = (uintptr_t) qp;
ret = ibv_cmd_create_qp(pd, &qp->ibv_qp, attr,
&cmd.ibv_cmd, sizeof(cmd),
diff --git a/providers/pvrdma/verbs.c b/providers/pvrdma/verbs.c
index f20ea6c..7dc9c83 100644
--- a/providers/pvrdma/verbs.c
+++ b/providers/pvrdma/verbs.c
@@ -79,7 +79,7 @@ int pvrdma_query_port(struct ibv_context *context, uint8_t port,
struct ibv_pd *pvrdma_alloc_pd(struct ibv_context *context)
{
struct ibv_alloc_pd cmd;
- struct pvrdma_alloc_pd_resp resp;
+ struct user_pvrdma_alloc_pd_resp resp;
struct pvrdma_pd *pd;
pd = malloc(sizeof(*pd));
@@ -92,7 +92,7 @@ struct ibv_pd *pvrdma_alloc_pd(struct ibv_context *context)
return NULL;
}
- pd->pdn = resp.pdn;
+ pd->pdn = resp.udata.pdn;
return &pd->ibv_pd;
}
--
1.8.5.6
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* Re: [PATCH v4 3/9] selinux lsm IB/core: Implement LSM notification system
From: Daniel Jurgens @ 2016-11-08 23:41 UTC (permalink / raw)
To: kbuild test robot
Cc: kbuild-all-JC7UmRfGjtg@public.gmane.org,
chrisw-69jw2NvuJkxg9hUCZPvPmw@public.gmane.org,
paul-r2n+y4ga6xFZroRs9YW3xA@public.gmane.org,
sds-+05T5uksL2qpZYMLLGbcSA@public.gmane.org,
eparis-FjpueFixGhCM4zKIHC2jIg@public.gmane.org,
dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org,
sean.hefty-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
hal.rosenstock-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org,
selinux-+05T5uksL2qpZYMLLGbcSA@public.gmane.org,
linux-security-module-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Yevgeny Petrilin, Liran Liss, Leon Romanovsky
In-Reply-To: <201611090600.BYNmwuAI%fengguang.wu@intel.com>
On 11/8/2016 4:36 PM, kbuild test robot wrote:
> Hi Daniel,
>
> [auto build test ERROR on rdma/master]
> [also build test ERROR on v4.9-rc4]
> [cannot apply to next-20161108]
> [if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
>
> url: https://github.com/0day-ci/linux/commits/Dan-Jurgens/SELinux-support-for-Infiniband-RDMA/20161109-053432
> base: https://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma.git master
> config: i386-randconfig-s1-201645 (attached as .config)
> compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
> reproduce:
> # save the attached .config to linux build tree
> make ARCH=i386
>
> All errors (new ones prefixed by >>):
>
>>> ERROR: "unregister_lsm_notifier" [drivers/infiniband/core/ib_core.ko] undefined!
>>> ERROR: "register_lsm_notifier" [drivers/infiniband/core/ib_core.ko] undefined!
> ---
> 0-DAY kernel test infrastructure Open Source Technology Center
> https://lists.01.org/pipermail/kbuild-all Intel Corporation
This link error happens when CONFIG_SECURITY is not set. I will address it in v5 after giving some time for additional comments.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH rdma-core 0/8] libpvrdma: userspace library for PVRDMA
From: Christoph Hellwig @ 2016-11-09 1:17 UTC (permalink / raw)
To: Adit Ranadive
Cc: Doug Ledford, Christoph Hellwig, Jason Gunthorpe,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, pv-drivers
In-Reply-To: <f67c81ff-d358-7242-dc52-e32dff252904-pghWNbHTmq7QT0dZR+AlfA@public.gmane.org>
FYI, the convention used in scsi is vmw_pvscsi.c, so naming the
RDMA equivalent vmw_pvrdma would make a lot of sense and still
be reasonably short.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH rdma-core 0/8] libpvrdma: userspace library for PVRDMA
From: Christoph Hellwig @ 2016-11-09 1:23 UTC (permalink / raw)
To: Jason Gunthorpe
Cc: Adit Ranadive, dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, pv-drivers
In-Reply-To: <20161105150136.GA23803-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
On Sat, Nov 05, 2016 at 09:01:36AM -0600, Jason Gunthorpe wrote:
> > Not entirely. You want me to keep the ABI file from the kernel in the
> > fix up folder and also keep a file with the modified structs in
> > providers/pvrdma?
>
> Yes, and the file with the modified structs should include the kernel
> header and duplicate it minimally. This will make it simpler for us to
> eventually get rid of it.
Can we just automate generating the user header, e.g. have a sed script
that recognized a magic comments ala
/* LIBIBVERBS PREAMBLE (DO NOT REMOVE) */
and just insers the needed fields?
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH v4 9/9] selinux: Add a cache for quicker retreival of PKey SIDs
From: kbuild test robot @ 2016-11-09 5:09 UTC (permalink / raw)
Cc: kbuild-all-JC7UmRfGjtg, chrisw-69jw2NvuJkxg9hUCZPvPmw,
paul-r2n+y4ga6xFZroRs9YW3xA, sds-+05T5uksL2qpZYMLLGbcSA,
eparis-FjpueFixGhCM4zKIHC2jIg, dledford-H+wXaHxf7aLQT0dZR+AlfA,
sean.hefty-ral2JQCrhuEAvxtiuMwx3w,
hal.rosenstock-Re5JQEeQqe8AvxtiuMwx3w,
selinux-+05T5uksL2qpZYMLLGbcSA,
linux-security-module-u79uwXL29TY76Z2rM5mHXA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA,
yevgenyp-VPRAkNaXOzVWk0Htik3J/w, liranl-VPRAkNaXOzVWk0Htik3J/w,
leonro-VPRAkNaXOzVWk0Htik3J/w, Daniel Jurgens
In-Reply-To: <1478639185-47521-10-git-send-email-danielj-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Hi Daniel,
[auto build test WARNING on rdma/master]
[also build test WARNING on v4.9-rc4]
[cannot apply to next-20161108]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
url: https://github.com/0day-ci/linux/commits/Dan-Jurgens/SELinux-support-for-Infiniband-RDMA/20161109-053432
base: https://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma.git master
reproduce:
# apt-get install sparse
make ARCH=x86_64 allmodconfig
make C=1 CF=-D__CHECK_ENDIAN__
sparse warnings: (new ones prefixed by >>)
include/linux/compiler.h:253:8: sparse: attribute 'no_sanitize_address': unknown attribute
>> security/selinux/ibpkey.c:116:24: sparse: incompatible types in comparison expression (different address spaces)
vim +116 security/selinux/ibpkey.c
100 * Description:
101 * Add a new pkey record to the hash table.
102 *
103 */
104 static void sel_pkey_insert(struct sel_pkey *pkey)
105 {
106 unsigned int idx;
107
108 /* we need to impose a limit on the growth of the hash table so check
109 * this bucket to make sure it is within the specified bounds
110 */
111 idx = sel_pkey_hashfn(pkey->psec.pkey);
112 list_add_rcu(&pkey->list, &sel_pkey_hash[idx].list);
113 if (sel_pkey_hash[idx].size == SEL_PKEY_HASH_BKT_LIMIT) {
114 struct sel_pkey *tail;
115
> 116 tail = list_entry(
117 rcu_dereference_protected(
118 sel_pkey_hash[idx].list.prev,
119 lockdep_is_held(&sel_pkey_lock)),
120 struct sel_pkey, list);
121 list_del_rcu(&tail->list);
122 kfree_rcu(tail, rcu);
123 } else {
124 sel_pkey_hash[idx].size++;
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: rdma-core example spec file is broken
From: Leon Romanovsky @ 2016-11-09 5:59 UTC (permalink / raw)
To: Alaa Hleihel
Cc: Jason Gunthorpe, linux-rdma-u79uwXL29TY76Z2rM5mHXA,
yishaih-VPRAkNaXOzVWk0Htik3J/w
In-Reply-To: <f807db01-c4c7-0f64-fe6b-476d02b686b3-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 374 bytes --]
On Tue, Nov 08, 2016 at 04:47:21PM +0200, Alaa Hleihel wrote:
> I tested this patch.
> It resolves the issue.
Thanks, applied.
>
> Thanks,
> Alaa
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox