Add comments on how the ring access functions are named and how they
are supposed to be used for producers and consumers. The functions are
also reordered so that the consumer functions are in the beginning and
the producer functions in the end, for easier reference. Put this in a
separate patch as the diff might look a little odd, but no
functionality has changed in this patch.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-12-git-send-email-magnus.karlsson@intel.com
There are two unnecessary READ_ONCE of descriptor data. These are not
needed since the data is written by the producer before it signals
that the data is available by incrementing the producer pointer. As the
access to this producer pointer is serialized and the consumer always
reads the descriptor after it has read and synchronized with the
producer counter, the write of the descriptor will have fully
completed and it does not matter if the consumer has any read tearing.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-11-git-send-email-magnus.karlsson@intel.com
Change the names of the validation functions to better reflect what
they are doing. The uppermost ones are reading entries from the rings
and only the bottom ones validate entries. So xskq_cons_read_ is a
better prefix name.
Also change the xskq_cons_read_ functions to return a bool
as the the descriptor or address is already returned by reference
in the parameters. Everyone is using the return value as a bool
anyway.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-9-git-send-email-magnus.karlsson@intel.com
Simplify and refactor consumer ring functions. The consumer first
"peeks" to find descriptors or addresses that are available to
read from the ring, then reads them and finally "releases" these
descriptors once it is done. The two local variables cons_tail
and cons_head are turned into one single variable called
cached_cons. cached_tail referred to the cached value of the
global consumer pointer and will be stored in cached_cons. For
cached_head, we just use cached_prod instead as it was not used
for a consumer queue before. It also better reflects what it
really is now: a cached copy of the producer pointer.
The names of the functions are also renamed in the same manner as
the producer functions. The new functions are called xskq_cons_
followed by what it does.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-8-git-send-email-magnus.karlsson@intel.com
At this point, there are no users of the functions xskq_nb_avail and
xskq_nb_free that take any other number of entries argument than 1, so
let us get rid of the second argument that takes the number of
entries.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-7-git-send-email-magnus.karlsson@intel.com
In the xsk consumer ring code there is a variable called RX_BATCH_SIZE
that dictates the minimum number of entries that we try to grab from
the fill and Tx rings. In fact, the code always try to grab the
maximum amount of entries from these rings. The only thing this
variable does is to throw an error if there is less than 16 (as it is
defined) entries on the ring. There is no reason to do this and it
will just lead to weird behavior from user space's point of view. So
eliminate this variable.
With this change, we will be able to simplify the xskq_nb_free and
xskq_nb_avail code in the next commit.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-6-git-send-email-magnus.karlsson@intel.com
Adopt the naming of the producer ring access functions to have a
similar naming convention as the functions in libbpf, but adapted to
the kernel. You first reserve a number of entries that you later
submit to the global state of the ring. This is much clearer, IMO,
than the one that was in the kernel part. Once renamed, we also
discover that two functions are actually the same, so remove one of
them. Some of the primitive ring submission operations are also the
same so break these out into __xskq_prod_submit that the upper level
ring access functions can use.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-5-git-send-email-magnus.karlsson@intel.com
Currently, the xsk ring code has two cached producer pointers:
prod_head and prod_tail. This patch consolidates these two into a
single one called cached_prod to make the code simpler and easier to
maintain. This will be in line with the user space part of the the
code found in libbpf, that only uses a single cached pointer.
The Rx path only uses the two top level functions
xskq_produce_batch_desc and xskq_produce_flush_desc and they both use
prod_head and never prod_tail. So just move them over to
cached_prod.
The Tx XDP_DRV path uses xskq_produce_addr_lazy and
xskq_produce_flush_addr_n and unnecessarily operates on both prod_tail
and prod_head, so move them over to just use cached_prod by skipping
the intermediate step of updating prod_tail.
The Tx path in XDP_SKB mode uses xskq_reserve_addr and
xskq_produce_addr. They currently use both cached pointers, but we can
operate on the global producer pointer in xskq_produce_addr since it
has to be updated anyway, thus eliminating the use of both cached
pointers. We can also remove the xskq_nb_free in xskq_produce_addr
since it is already called in xskq_reserve_addr. No need to do it
twice.
When there is only one cached producer pointer, we can also simplify
xskq_nb_free by removing one argument.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-4-git-send-email-magnus.karlsson@intel.com
In order to set the correct return flags for poll, the xsk code has to
check if the Rx queue is empty and if the Tx queue is full. This code
was unnecessarily large and complex as it used the functions that are
used to update the local state from the global state (xskq_nb_free and
xskq_nb_avail). Since we are not doing this nor updating any data
dependent on this state, we can simplify the functions. Another
benefit from this is that we can also simplify the xskq_nb_free and
xskq_nb_avail functions in a later commit.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-3-git-send-email-magnus.karlsson@intel.com
The lazy update threshold was introduced to keep the producer and
consumer some distance apart in the completion ring. This was
important in the beginning of the development of AF_XDP as the ring
format as that point in time was very sensitive to the producer and
consumer being on the same cache line. This is not the case
anymore as the current ring format does not degrade in any noticeable
way when this happens. Moreover, this threshold makes it impossible
to run the system with rings that have less than 128 entries.
So let us remove this threshold and just get one entry from the ring
as in all other functions. This will enable us to remove this function
in a later commit. Note that xskq_produce_addr_lazy followed by
xskq_produce_flush_addr_n are still not the same function as
xskq_produce_addr() as it operates on another cached pointer.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/1576759171-28550-2-git-send-email-magnus.karlsson@intel.com
Currently, addresses are chunk size aligned. This means, we are very
restricted in terms of where we can place chunk within the umem. For
example, if we have a chunk size of 2k, then our chunks can only be placed
at 0,2k,4k,6k,8k... and so on (ie. every 2k starting from 0).
This patch introduces the ability to use unaligned chunks. With these
changes, we are no longer bound to having to place chunks at a 2k (or
whatever your chunk size is) interval. Since we are no longer dealing with
aligned chunks, they can now cross page boundaries. Checks for page
contiguity have been added in order to keep track of which pages are
followed by a physically contiguous page.
Signed-off-by: Kevin Laatz <kevin.laatz@intel.com>
Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
This commit adds support for a new flag called need_wakeup in the
AF_XDP Tx and fill rings. When this flag is set, it means that the
application has to explicitly wake up the kernel Rx (for the bit in
the fill ring) or kernel Tx (for bit in the Tx ring) processing by
issuing a syscall. Poll() can wake up both depending on the flags
submitted and sendto() will wake up tx processing only.
The main reason for introducing this new flag is to be able to
efficiently support the case when application and driver is executing
on the same core. Previously, the driver was just busy-spinning on the
fill ring if it ran out of buffers in the HW and there were none on
the fill ring. This approach works when the application is running on
another core as it can replenish the fill ring while the driver is
busy-spinning. Though, this is a lousy approach if both of them are
running on the same core as the probability of the fill ring getting
more entries when the driver is busy-spinning is zero. With this new
feature the driver now sets the need_wakeup flag and returns to the
application. The application can then replenish the fill queue and
then explicitly wake up the Rx processing in the kernel using the
syscall poll(). For Tx, the flag is only set to one if the driver has
no outstanding Tx completion interrupts. If it has some, the flag is
zero as it will be woken up by a completion interrupt anyway.
As a nice side effect, this new flag also improves the performance of
the case where application and driver are running on two different
cores as it reduces the number of syscalls to the kernel. The kernel
tells user space if it needs to be woken up by a syscall, and this
eliminates many of the syscalls.
This flag needs some simple driver support. If the driver does not
support this, the Rx flag is always zero and the Tx flag is always
one. This makes any application relying on this feature default to the
old behaviour of not requiring any syscalls in the Rx path and always
having to call sendto() in the Tx path.
For backwards compatibility reasons, this feature has to be explicitly
turned on using a new bind flag (XDP_USE_NEED_WAKEUP). I recommend
that you always turn it on as it so far always have had a positive
performance impact.
The name and inspiration of the flag has been taken from io_uring by
Jens Axboe. Details about this feature in io_uring can be found in
http://kernel.dk/io_uring.pdf, section 8.3.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Add a function that checks whether the Fill Ring has the specified
amount of descriptors available. It will be useful for mlx5e that wants
to check in advance, whether it can allocate a bulk of RX descriptors,
to get the best performance.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Acked-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Clang warns:
In file included from net/xdp/xsk_queue.c:10:
net/xdp/xsk_queue.h:292:2: warning: expression result unused
[-Wunused-value]
WRITE_ONCE(q->ring->producer, q->prod_tail);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
include/linux/compiler.h:284:6: note: expanded from macro 'WRITE_ONCE'
__u.__val; \
~~~ ^~~~~
1 warning generated.
The q->prod_tail assignment has a comma at the end, not a semi-colon.
Fix that so clang no longer warns and everything works as expected.
Fixes: c497176cb2 ("xsk: add Rx receive functions and poll support")
Link: https://github.com/ClangBuiltLinux/linux/issues/544
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Acked-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Acked-by: Björn Töpel <bjorn.topel@intel.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The ring buffer code of XDP sockets is missing a memory barrier on the
consumer side between the load of the data and the write that signals
that it is ok for the producer to put new data into the buffer. On
architectures that does not guarantee that stores are not reordered
with older loads, the producer might put data into the ring before the
consumer had the chance to read it. As IA does guarantee this
ordering, it would only need a compiler barrier here, but there are no
primitives in Linux for this specific case (hinder writes to be ordered
before older reads) so I had to add a smp_mb() here which will
translate into a run-time synch operation on IA.
Added a longish comment in the code explaining what each barrier in
the ring implementation accomplishes and what would happen if we
removed one of them.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Passing a non-existing option in the options member of struct
xdp_desc was, incorrectly, silently ignored. This patch addresses
that behavior, and drops any Tx descriptor with non-existing options.
We have examined existing user space code, and to our best knowledge,
no one is relying on the current incorrect behavior. AF_XDP is still
in its infancy, so from our perspective, the risk of breakage is very
low, and addressing this problem now is important.
Fixes: 35fcde7f8d ("xsk: support for Tx")
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
XSK UMEM is strongly single producer single consumer so reuse of
frames is challenging. Add a simple "stash" of FILL packets to
reuse for drivers to optionally make use of. This is useful
when driver has to free (ndo_stop) or resize a ring with an active
AF_XDP ZC socket.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This commit gets rid of the structure xdp_umem_props. It was there to
be able to break a dependency at one point, but this is no longer
needed. The values in the struct are instead stored directly in the
xdp_umem structure. This simplifies the xsk code as well as af_xdp
zero-copy drivers and as a bonus gets rid of one internal header file.
The i40e driver is also adapted to the new interface in this commit.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Polling for the ingress queues relies on reading the producer/consumer
pointers of the Rx queue.
Prior this commit, a cached consumer pointer could be used, instead of
the actual consumer pointer and therefore report POLLIN prematurely.
This patch makes sure that the non-cached consumer pointer is used
instead.
Reported-by: Qi Zhang <qi.z.zhang@intel.com>
Tested-by: Qi Zhang <qi.z.zhang@intel.com>
Fixes: c497176cb2 ("xsk: add Rx receive functions and poll support")
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The code in xskq_produce_addr erroneously checked if there
was up to LAZY_UPDATE_THRESHOLD amount of space in the completion
queue. It only needs to check if there is one slot left in the
queue. This bug could under some circumstances lead to a WARN_ON_ONCE
being triggered and the completion message to user space being lost.
Fixes: 35fcde7f8d ("xsk: support for Tx")
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Reported-by: Pavel Odintsov <pavel@fastnetmon.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Here we add the functionality required to support zero-copy Tx, and
also exposes various zero-copy related functions for the netdevs.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Moved struct xdp_umem to xdp_sock.h, in order to prepare for zero-copy
support.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Currently, AF_XDP only supports a fixed frame-size memory scheme where
each frame is referenced via an index (idx). A user passes the frame
index to the kernel, and the kernel acts upon the data. Some NICs,
however, do not have a fixed frame-size model, instead they have a
model where a memory window is passed to the hardware and multiple
frames are filled into that window (referred to as the "type-writer"
model).
By changing the descriptor format from the current frame index
addressing scheme, AF_XDP can in the future be extended to support
these kinds of NICs.
In the index-based model, an idx refers to a frame of size
frame_size. Addressing a frame in the UMEM is done by offseting the
UMEM starting address by a global offset, idx * frame_size + offset.
Communicating via the fill- and completion-rings are done by means of
idx.
In this commit, the idx is removed in favor of an address (addr),
which is a relative address ranging over the UMEM. To convert an
idx-based address to the new addr is simply: addr = idx * frame_size +
offset.
We also stop referring to the UMEM "frame" as a frame. Instead it is
simply called a chunk.
To transfer ownership of a chunk to the kernel, the addr of the chunk
is passed in the fill-ring. Note, that the kernel will mask addr to
make it chunk aligned, so there is no need for userspace to do
that. E.g., for a chunk size of 2k, passing an addr of 2048, 2050 or
3000 to the fill-ring will refer to the same chunk.
On the completion-ring, the addr will match that of the Tx descriptor,
passed to the kernel.
Changing the descriptor format to use chunks/addr will allow for
future changes to move to a type-writer based model, where multiple
frames can reside in one chunk. In this model passing one single chunk
into the fill-ring, would potentially result in multiple Rx
descriptors.
This commit changes the uapi of AF_XDP sockets, and updates the
documentation.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Previously the fill queue descriptor was not copied to kernel space
prior validating it, making it possible for userland to change the
descriptor post-kernel-validation.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
In this commit we remove the explicit ring structure from the the
uapi. It is tricky for an uapi to depend on a certain L1 cache line
size, since it can differ for variants of the same architecture. Now,
we let the user application determine the offsets of the producer,
consumer and descriptors by asking the socket via getsockopt.
A typical flow would be (Rx ring):
struct xdp_mmap_offsets off;
struct xdp_desc *ring;
u32 *prod, *cons;
void *map;
...
getsockopt(fd, SOL_XDP, XDP_MMAP_OFFSETS, &off, &optlen);
map = mmap(NULL, off.rx.desc +
NUM_DESCS * sizeof(struct xdp_desc),
PROT_READ | PROT_WRITE,
MAP_SHARED | MAP_POPULATE, sfd,
XDP_PGOFF_RX_RING);
prod = map + off.rx.producer;
cons = map + off.rx.consumer;
ring = map + off.rx.desc;
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Removed some cases of unnecessary parentheses.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Clean up SPDX-License-Identifier and removing licensing leftovers.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
In this commit, a new getsockopt is added: XDP_STATISTICS. This is
used to obtain stats from the sockets.
v2: getsockopt now returns size of stats structure.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Here, Tx support is added. The user fills the Tx queue with frames to
be sent by the kernel, and let's the kernel know using the sendmsg
syscall.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Here the actual receive functions of AF_XDP are implemented, that in a
later commit, will be called from the XDP layers.
There's one set of functions for the XDP_DRV side and another for
XDP_SKB (generic).
A new XDP API, xdp_return_buff, is also introduced.
Adding xdp_return_buff, which is analogous to xdp_return_frame, but
acts upon an struct xdp_buff. The API will be used by AF_XDP in future
commits.
Support for the poll syscall is also implemented.
v2: xskq_validate_id did not update cons_tail.
The entries variable was calculated twice in xskq_nb_avail.
Squashed xdp_return_buff commit.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Here, the bind syscall is added. Binding an AF_XDP socket, means
associating the socket to an umem, a netdev and a queue index. This
can be done in two ways.
The first way, creating a "socket from scratch". Create the umem using
the XDP_UMEM_REG setsockopt and an associated fill queue with
XDP_UMEM_FILL_QUEUE. Create the Rx queue using the XDP_RX_QUEUE
setsockopt. Call bind passing ifindex and queue index ("channel" in
ethtool speak).
The second way to bind a socket, is simply skipping the
umem/netdev/queue index, and passing another already setup AF_XDP
socket. The new socket will then have the same umem/netdev/queue index
as the parent so it will share the same umem. You must also set the
flags field in the socket address to XDP_SHARED_UMEM.
v2: Use PTR_ERR instead of passing error variable explicitly.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Another setsockopt (XDP_RX_QUEUE) is added to let the process allocate
a queue, where the kernel can pass completed Rx frames from the kernel
to user process.
The mmapping of the queue is done using the XDP_PGOFF_RX_QUEUE offset.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Here, we add another setsockopt for registered user memory (umem)
called XDP_UMEM_FILL_QUEUE. Using this socket option, the process can
ask the kernel to allocate a queue (ring buffer) and also mmap it
(XDP_UMEM_PGOFF_FILL_QUEUE) into the process.
The queue is used to explicitly pass ownership of umem frames from the
user process to the kernel. These frames will in a later patch be
filled in with Rx packet data by the kernel.
v2: Fixed potential crash in xsk_mmap.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Magnus Karlsson