An earlier patch which introduced smp_load_acquire/smp_store_release
into rxe_queue.h incorrectly assumed that surrounding spin-locks in
rxe_verbs.c around queue updates for kernel ulps was sufficient to
protect the passing of data through the queues between the ulp and
the rxe tasklets. But this was incorrect. The typical sequence was
ulp rxe requester tasklet
------------------------ ---------------------
spin_lock_irqsave() wqe = queue_head(queue)
if (!queue_full(q)) { if (!wqe)
spin_unlock_irqrestore return;
return -ENOMEM
} <process wqe>
wqe = queue_producer_addr(q)
<fill in wqe> queue_advance_consumer(queue)
queue_advance_producer(q)
spin_unlock_irqrestore()
queue_head() calls queue_empty() which calls smp_load_acquire()
For user space apps queue_advance_producer() calls smp_store_release()
so that there is a memory barrier between the producer and the
consumer but for kernel ulps queue_advance_produce() just incremented
the producer index because the lock function is a release function.
But to work the barrier has to come between filling in the wqe and
updating the producer index. This patch adds the missing barriers.
It also changes the enum names for the ulp queue types to
QUEUE_TYPE_FROM/TO_ULP instead of QUEUE_TYPE_TO/FROM_DRIVER
which is very ambiguous. This bug is suspected as the cause of very
rare lockups in a very high scale storage application. It is a bug
in any case and should be corrected.
Fixes: 0a67c46d2e ("RDMA/rxe: Protect user space index loads/stores")
Link: https://lore.kernel.org/r/20230214071053.5395-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
rxe_queue.h currently includes <uapi/rdma/rdma_user_rxe.h> for a
definition of struct rxe_queue_buf. But it is only used as a pointer so
the definition is not needed.
This patch replaces the include statement with the declaration
struct rxe_queue_buf;
Link: https://lore.kernel.org/r/20220630190425.2251-5-rpearsonhpe@gmail.com
Reported-by: Frank Zago <frank.zago@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Earlier patches added memory barriers to protect user space to kernel
space communications. The user space queues were previously shown to have
occasional memory synchonization errors which were removed by adding
smp_load_acquire, smp_store_release barriers. This patch extends that to
the case where queues are used between kernel space threads.
This patch also extends the queue types to include kernel ULP queues which
access the other end of the queues in kernel verbs calls like poll_cq and
post_send/recv.
Link: https://lore.kernel.org/r/20210914164206.19768-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
In order to prevent user space from modifying the index that belongs to
the kernel for shared queues let the kernel use a local copy of the index
and copy any new values of that index to the shared rxe_queue_bus struct.
This adds more switch statements which decreases the performance of the
queue API. Move the type into the parameter list for these functions so
that the compiler can optimize out the switch statements when the explicit
type is known. Modify all the calls in the driver on performance paths to
pass in the explicit queue type.
Link: https://lore.kernel.org/r/20210527194748.662636-4-rpearsonhpe@gmail.com
Link: https://lore.kernel.org/linux-rdma/20210526165239.GP1002214@@nvidia.com/
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Modify the queue APIs to protect all user space index loads with
smp_load_acquire() and all user space index stores with
smp_store_release(). Base this on the types of the queues which can be one
of ..KERNEL, ..FROM_USER, ..TO_USER. Kernel space indices are protected by
locks which also provide memory barriers.
Link: https://lore.kernel.org/r/20210527194748.662636-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
To create optimal code only want to use smp_load_acquire() and
smp_store_release() for user indices in rxe_queue APIs since kernel
indices are protected by locks which also act as memory barriers. By
adding a type to the queues we can determine which indices need to be
protected.
Link: https://lore.kernel.org/r/20210527194748.662636-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Change work and completion queues to use smp_load_acquire() and
smp_store_release() to synchronize between driver and users. This commit
goes with a matching series of commits in the rxe user space provider.
Link: https://lore.kernel.org/r/20201210174258.5234-1-rpearson@hpe.com
Signed-off-by: Bob Pearson <rpearson@hpe.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 7649773293 ("cxgb3/l2t: Fix undefined behaviour")
Link: https://lore.kernel.org/r/20200213010425.GA13068@embeddedor.com
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> # added a few more
Now when ib_udata is passed to all the driver's object create/destroy APIs
the ib_udata will carry the ib_ucontext for every user command. There is
no need to also pass the ib_ucontext via the functions prototypes.
Make ib_udata the only argument psssed.
Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Open coding pointer math is not acceptable for describing the uABI in
RDMA. Provide structs for all the cases.
The udata is casted to the struct as close to the verbs entry point
as possible for maximum clarity. Function signatures and so forth
are revised to allow for this.
Tested-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
RXE resets the send-q only once in rxe_qp_init_req() when
QP is created, but when the QP is reused after QP reset, the send-q
holds previous garbage data.
This garbage data wrongly fails CQEs that otherwise
should have completed successfully.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Signed-off-by: Yonatan Cohen <yonatanc@mellanox.com>
Reviewed-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Soft RoCE (RXE) - The software RoCE driver
ib_rxe implements the RDMA transport and registers to the RDMA core
device as a kernel verbs provider. It also implements the packet IO
layer. On the other hand ib_rxe registers to the Linux netdev stack
as a udp encapsulating protocol, in that case RDMA, for sending and
receiving packets over any Ethernet device. This yields a RDMA
transport over the UDP/Ethernet network layer forming a RoCEv2
compatible device.
The configuration procedure of the Soft RoCE drivers requires
binding to any existing Ethernet network device. This is done with
/sys interface.
A userspace Soft RoCE library (librxe) provides user applications
the ability to run with Soft RoCE devices. The use of rxe verbs ins
user space requires the inclusion of librxe as a device specifics
plug-in to libibverbs. librxe is packaged separately.
Architecture:
+-----------------------------------------------------------+
| Application |
+-----------------------------------------------------------+
+-----------------------------------+
| libibverbs |
User +-----------------------------------+
+----------------+ +----------------+
| librxe | | HW RoCE lib |
+----------------+ +----------------+
+---------------------------------------------------------------+
+--------------+ +------------+
| Sockets | | RDMA ULP |
+--------------+ +------------+
+--------------+ +---------------------+
| TCP/IP | | ib_core |
+--------------+ +---------------------+
+------------+ +----------------+
Kernel | ib_rxe | | HW RoCE driver |
+------------+ +----------------+
+------------------------------------+
| NIC driver |
+------------------------------------+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+-----------------------------------------------------------+
| Application |
+-----------------------------------------------------------+
+-----------------------------------+
| libibverbs |
User +-----------------------------------+
+----------------+ +----------------+
| librxe | | HW RoCE lib |
+----------------+ +----------------+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+--------------+ +------------+
| Sockets | | RDMA ULP |
+--------------+ +------------+
+--------------+ +---------------------+
| TCP/IP | | ib_core |
+--------------+ +---------------------+
+------------+ +----------------+
Kernel | ib_rxe | | HW RoCE driver |
+------------+ +----------------+
+------------------------------------+
| NIC driver |
+------------------------------------+
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Soft RoCE resources:
[1[ https://github.com/SoftRoCE/librxe-dev librxe - source code in
Github
[2] https://github.com/SoftRoCE/rxe-dev/wiki/rxe-dev:-Home - Soft RoCE
Wiki page
[3] https://github.com/SoftRoCE/librxe-dev - Soft RoCE userspace library
Signed-off-by: Kamal Heib <kamalh@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: Moni Shoua <monis@mellanox.com>
Reviewed-by: Haggai Eran <haggaie@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
