This manual page describes the zerocopy interface of RDS, which
was added in RDSv3. For a description of the basic RDS interface,
please refer to
The principal mode of operation for RDS zerocopy is like this:
one participant (the client) wishes to initiate a direct transfer
to or from some area of memory in its process address space.
This memory does not have to be aligned.
The client obtains a handle for this region of memory, and
passes it to the other participant (the server). This is called
the RDMA cookie. To the application, the cookie is an opaque 64bit
The client sends this handle to
the server application, along with other details of the RDMA
request (such as which data to transfer to that memory area).
Throughout the following discussion, we will refer to this
message as the RDMA request.
The server uses this RDMA cookie to initiate the requested RDMA
transfer. The RDMA transfer is combined atomically with a
normal RDS message, which is delivered to the client. This
message is called the RDMA ACK throughout the following. Atomic
in this context means that either both the RDMA succeeds and the
RDMA ACK is delivered, or neither succeeds.
Thus, when the client receives the RDMA ACK, it knows that
the RDMA has completed successfully. It can then release the
RDMA cookie for this memory region, if it wishes to.
RDMA operations are not reliable, in the sense that unlike normal
RDS messages, RDS RDMA operations may fail, and get
The interface is currently based on control messages (ancillary
data) sent or received via the
system calls. Optionally, an older interface can be used that
is based on the
system call. However, we recommend using control messages, as
this reduces the number of system calls required.
Control message interface
With the control message interface, the RDMA cookie is passed to
the server out-of-band, included in an extension header attached
to the RDS message.
The following outlines the mode of operation; the data
types used will be specified in details in a subsequent section.
Initially, the client will send RDMA requests along with a
control message. The control message contains the address and
length of the memory region for which to obtain a handle, some
flags, and a pointer to a memory location (in the caller's address
space) where the kernel will store the RDMA cookie.
Alternatively, if the application has already obtained a RDMA cookie
for the memory range it wants to RDMA to/from, it can hand this
cookie to the kernel using the
Either way, the kernel will include the resulting RDMA cookie
in an extension header that is transmitted as part of the RDMA
request to the server.
When the server receives the RDMA request, the kernel will deliver the
cookie wrapped inside a
The server then initiates the data transfer by sending the RDMA ACK message
along with a
control message. This message contains the RDMA cookie, and the local
memory to copy to or from.
The server process may request a notification when an RDMA operation
completes. Notifications are delivered as a
control messages. When an application calls
it will either receive a regular RDS message (possibly with other RDMA
related control messages), or an empty message with one or more
status control messages.
In addition, applications
When an RDMA operation fails for some reason and is discarded, the
application can ask to receive notifications for failed messages as
well, regardless of whether it asked for success notification of an
individual message or not. This behavior is turned on by setting the
In addition to the control message interface, RDS allows a process
to register and release memory ranges for RDMA through calls to
To obtain a RDMA cookie for a given memory range, the application can
setsockopt with RDS_GET_MR.
This operates essentially the same way as the
control message: the argument contains the address and length of the
memory range to be registered, and a pointer to a RDMA cookie variable,
in which the system call will store the cookie for the registered
Memory ranges can be released by calling
setsockopt with RDS_FREE_MR,
giving the RDMA cookie and additional flags as arguments.
This is a boolean option which can be set as well as queried
When enabled, RDS will send RDMA notification messages to
the application for any RDMA operation that fails. This
option defaults to off.
For all of these calls, the
RDMA MACROS AND TYPES
typedef u_int64_t rds_rdma_cookie_t
This encapsulates a memory location in the client process. In the
current implementation, it contains the R_Key of the remote memory
region, and the offset into it (so that the application does not
have to worry about alignment.
The RDMA cookie is used in several struct types described below.
control message contains a rds_rdma_cookie_t all by itself as payload.
The following data type is used with
control messages and with the
specifies a memory location where to store the RDMA cookie.
value is a bitwise OR of any of the following flags:
This tells the kernel that the allocated RDMA cookie is to be used
exactly once. When the RDMA ACK message arrives, the kernel will
automatically unbind the memory area and release any resources
associated with the cookie.
If this flag is not set, it is the application's responsibility to
release the memory region at a later time using the
Normally, RDMA memory mappings are invalidated lazily, as this
requires some relatively costly synchronization with the HCA. However,
this means that the server application can continue to access the
registered memory for some indeterminate amount of time.
If this flag is set, the RDS code will invalidate
the mapping at the time it is released (either upon arrival of the
RDMA ACK, if
was specified; or when the application destroys it using
RDMA operations are initiated by the server using the
control message, which takes the following data as payload:
argument contains the RDMA cookie received from the client.
The local memory is given via an array of
The array address is given in
and its number of elements is given in
The struct member
specifies a location relative to the memory area identified
by the cookie:
is an offset into that region, and
is the length of the memory window to copy to/from.
This length must match the size of the local memory area,
i.e. the sum of bytes in all members of the local iovec.
The flags field contains the bitwise OR of any of the following
If set, any RDMA WRITE is initiated from the server's memory
to the client's. If not set, RDS will do a RDMA READ from the
client's memory to the server's memory.
By default, Infiniband makes no guarantee about the ordering of
an RDMA READ with respect to subsequent SEND operations. Setting
this flag asks that the RDMA READ should be fenced off the subsequent
RDS ACK message. Setting this flag requires an additional round-trip
of the IB fabric, but it is a good idea to use set this flag
by default, unless you are really sure you do not want it.
This flag requests a notification upon completion of the RDMA
operation (successful or otherwise). The noticiation will contain
the value of the
field passed in by the application. This allows the application to
release resources (such as buffers) assosicated with the RDMA transfer.
can be used to pass an application specific identifier to the
kernel. This token is returned to the application when a status
notification is generated (see the following section).
The RDS kernel code is able to notify the server application when
an RDMA operation completes. These notifications are delivered
By default, no notifications are generated. There are two ways an
application can request them. On one hand, status notifications can
be enabled on a per-operation basis by setting the
flag in the RDMA arguments. On the other hand, the application can
request notifications for all RDMA operations that fail by setting
socket option (see below).
In both cases, the format of the notification is the same; and at
most one notification will be sent per completed operation.
field contains the value previously given to the kernel in the
control message. The
field contains a status value, with 0 indicating success, and
non-zero indicating an error.
The following status codes are currently defined:
The RDMA operation succeeded.
The RDMA operation failed due to a remote access error. This is
usually due to an invalid R_key, offset or transfer size.
The RDMA operation was canceled by the application.
(This error code is not yet generated).
RDMA operations were discarded after the connection broke and
was re-established. The RDMA operation may have been processed
Any other failure.
RDMA setsockopt arguments
When using the
socket option to register a memory range, the application passes
a pointer to a
variable, described above.
call takes an argument of type
specifies the RDMA cookie to be released. RDMA access to the memory
range will usually not be invoked instantly, because the operation is
rather costly. However, if the
RDS will invalidate the indicated mapping immediately,
as described in section
argument is 0, and
is set, RDS will invalidate old memory mappings on all devices.
In addition to the usual error codes returned by
sendmsg, recvmsg and setsockopt,
RDS returns the following error codes:
RDS was unable to map a memory range because the limit was
exceeded (returned by
RDS_CMSG_RDMA_MAP and RDS_GET_MR).
When sending a message, there were were conflicting control messages
messages, or a
RDMA_MAP and a RDMA_DEST
RDS_CMSG_RDMA_MAP or RDS_GET_MR
operation, the application specified memory range greater than the
maximum size supported.
When setting up an RDMA operation with
the size of the local memory (given in the
did not match the size of the remote memory range.
RDS was unable to obtain a DMA mapping for the indicated memory.
Currently, the following limits apply
The maximum size of a zerocopy transfer is 1MB. This can be
adjusted via the
The maximum number of memory ranges that can be mapped is
limited to 2048 at the moment. This can be adjusted via the
module parameter. However, the actual limit imposed by the
hardware may in fact be lower.
RDS was written and is Copyright (C) 2007-2008 by Oracle, Inc.