Age | Commit message (Collapse) | Author | Files | Lines |
|
When XSK frame size is 3072 (or another power of two multiplied by 3),
KLM mechanism for NIC virtual memory page mapping can be optimized by
replacing it with KSM.
Before this change, two KLM entries were needed to map an XSK frame that
is not a power of two: one entry maps the UMEM memory up to the frame
length, the other maps the rest of the stride to the garbage page.
When the frame length divided by 3 is a power of two, it can be mapped
using 3 KSM entries, and the fourth will map the rest of the stride to
the garbage page. All 4 KSM entries are of the same size, which allows
for a much faster lookup.
Frame size 3072 is useful in certain use cases, because it allows
packing 4 frames into 3 pages. Generally speaking, other frame sizes
equal to PAGE_SIZE minus a power of two can be optimized in a similar
way, but it will require many more KSMs per frame, which slows down UMRs
a little bit, but more importantly may hit the limit for the maximum
number of KSM entries.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
XSK RQs support striding RQ linear mode, but the stride size may be
bigger than the XSK frame size, because:
1. The stride size must be a power of two.
2. The stride size must be equal to the UMR page size. Each XSK frame is
treated as a separate page, because they aren't necessarily adjacent in
physical memory, so the driver can't put more than one stride per page.
3. The minimal MTT page size is 4096 on older firmware.
That means that if XSK frame size is 2048 or not a power of two, the
strides may be bigger than XSK frames. Normally, it's not a problem if
the hardware enforces the MTU. However, traffic between vports skips the
hardware MTU check, and oversized packets may be received.
If an oversized packet is bigger than the XSK frame but not bigger than
the stride, it will cause overwriting of the adjacent UMEM region. If
the packet takes more than one stride, they can be recycled for reuse,
so it's not a problem when the XSK frame size matches the stride size.
Work around the above issue by leveraging KLM to make a more
fine-grained mapping. The beginning of each stride is mapped to the
frame memory, and the padding up to the closest power of two is mapped
to the overflow page that doesn't belong to UMEM. This way, application
data corruption won't happen upon receiving packets bigger than MTU.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Make mlx5e_mpwrq_mtts_per_wqe take into account that KSM requires
smaller alignment than MTT.
Ensure that there is always an even amount of MTTs in a UMR WQE, so that
complete octwords are formed, and no garbage is mapped.
Drop extra alignment in MLX5_MTT_OCTW that may cause setting too big
ucseg->xlt_octowords, also leading to mapping garbage.
Generalize some calculations by introducing the MLX5_OCTWORD constant.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Instead of passing the unaligned flag, pass an enum that indicates the
UMR mode. The next commit will add the third mode (KLM for certain
configurations of XSK), which will be added to this enum instead of
adding another bool flag everywhere.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The regular RQ remains open after opening an XSK socket, in order to
guarantee that closing the XSK socket never fails due to an error when
reopening the regular RQ.
To save memory, the regular RQ can be deactivated and flushed, releasing
all pages, when an XSK socket is open.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In the initial implementation of XSK in mlx5e, XSK RQs coexisted with
regular RQs in the same channel. The main idea was to allow RSS work the
same for regular traffic, without need to reconfigure RSS to exclude XSK
queues.
However, this scheme didn't prove to be beneficial, mainly because of
incompatibility with other vendors. Some tools don't properly support
using higher indices for XSK queues, some tools get confused with the
double amount of RQs exposed in sysfs. Some use cases are purely XSK,
and allocating the same amount of unused regular RQs is a waste of
resources.
This commit changes the queuing scheme to the standard one, where XSK
RQs replace regular RQs on the channels where XSK sockets are open. Two
RQs still exist in the channel to allow failsafe disable of XSK, but
only one is exposed at a time. The next commit will achieve the desired
memory save by flushing the buffers when the regular RQ is unused.
As the result of this transition:
1. It's possible to use RSS contexts over XSK RQs.
2. It's possible to dedicate all queues to XSK.
3. When XSK RQs coexist with regular RQs, the admin should make sure no
unwanted traffic goes into XSK RQs by either excluding them from RSS or
settings up the XDP program to return XDP_PASS for non-XSK traffic.
4. When using a mixed fleet of mlx5e devices and other netdevs, the same
configuration can be applied. If the application supports the fallback
to copy mode on unsupported drivers, it will work too.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Add a function to flush an RQ: clean up descriptors, release pages and
reset the RQ. This procedure is used by the recovery flow, and it will
also be used in a following commit to free some memory when switching a
channel to the XSK mode.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When fragments of different WQEs share the same page, mlx5e_post_rx_wqes
must wait until the old WQE stops using the page, only then the new WQE
can allocate the new page. Essentially, it means that if WQE index i is
still in use, the allocation must stop before `i % bulk`, where bulk is
the number of WQEs that may share the same page.
As bulk is always a power of two, `i % bulk = i & (bulk - 1)`, and the
new wqe_index_mask field will be equal to `bulk - 1`.
At the same time, wqe_bulk remains for optimization purposes and stores
`max(bulk, 8)`, which allows to skip the allocation until we have at
least 8 WQEs free.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
struct mlx5e_alloc_unit consists of a single union. Convert it to a
union itself to simplify casting it to struct xdp_buff *, which will be
used to implement XSK batching on striding RQ.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
mlx5e_alloc_unit stores the DMA address and a pointer to either struct
page (regular RQ) or struct xdp_buff (XSK RQ). This DMA address is
redundant, because when a page or an XSK frame is allocated, the same
address is also stored there. Some flows take the address from struct
mlx5e_alloc_unit, and some take it from struct page or xdp_buff.
This commit removes the address from struct mlx5e_alloc_unit, which
makes it twice as small and improves locality (this struct is used in an
array), also saving on unnecessary stores to the addr field. Almost all
flows know unambiguously whether the DMA address should be taken from
page or from xdp_buff. The exception is the allocation flows, where a
new branch appeared, which will be optimized out in the next commits.
struct mlx5e_alloc_unit used to be called mlx5e_dma_info.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The next commit will remove the DMA address from the struct currently
called mlx5e_dma_info, because the same value can be retrieved with
page_pool_get_dma_addr(page) in almost all cases, with the notable
exception of SHAMPO (HW GRO implementation) that modifies this address
on the fly, after the initial allocation.
To keep the SHAMPO logic intact, struct mlx5e_dma_info remains in the
SHAMPO code, consisting of addr and page (XSK is not compatible with
SHAMPO). The struct used in all other places is renamed to
mlx5e_alloc_unit, allowing the next commit to remove the addr field
without affecting SHAMPO.
The new name means "allocation unit", and it's more appropriate after
the field with the DMA address gets removed.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
RX page cache stores dma_info structs, that consist of a pointer to
struct page and a DMA address. In fact, the DMA address is extracted
from struct page using page_pool_get_dma_addr when a page is pushed to
the cache. By moving this call to the point when a page is popped from
the cache, we can avoid storing the DMA address in the cache,
effectively reducing its size by two times without losing any
functionality.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
UMR MTTs used in striding RQ have certain alignment requirements. While
it's guaranteed to work when UMR pages are aligned to the UMR page size,
in practice it works then UMR pages are aligned to 8 bytes. However,
it's still not enough flexibility for the unaligned mode of XSK. This
patch leverages KSM to map UMR pages without alignment requirements,
when unaligned XSK is active. The downside is that KSM entries are twice
as big as MTTs, which limits the maximum WQE size, so regular RQs and
aligned XSK continue using MTTs.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Currently, rq->mkey_be keeps a big-endian value of either the PA MKey
(for legacy RQ, no address translation) or MTT MKey (for striding RQ,
direct address translation). Striding RQ stores the same value in
rq->umr_mkey in the native endianness.
The next commit will make striding RQ use KSM MKey (indirect address
translation) for the unaligned mode of XSK, which will require storing
both KSM MKey and PA MKey in the RQ struct. This commit optimizes fields
of mlx5e_rq: umr_mkey is removed (it's redundant), mkey_be always points
to the PA MKey, and mpwqe.umr_mkey_be points to the MTT MKey (or to the
KSM MKey, starting from the next commit).
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This commit allows striding RQ to determine MTT page size at runtime,
instead of sticking to the compile-time PAGE_SIZE. This functionality
will be used by a following commit that adjusts the MTT page size to the
XSK frame size.
Stick with PAGE_SIZE for XSK on legacy RQ, as frag_stride is not used in
data path, it only helps calculate how pages are partitioned into
fragments, and PAGE_SIZE will ensure each fragment starts at the
beginning of a new allocation unit (XSK frame).
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Some of the parameters of striding RQ are compile-time constants, but
they are going to become dynamically calculated at runtime in a
following commit. This commit prepares the datapath to take cached
runtime parameters, prefilled at queue creation.
New fields added to struct mlx5e_rq fit into an existing 7-byte hole.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This commit moves the dma_info array to the end of struct mlx5e_mpw_info
to make it a flexible array. It also removes the intermediate struct
mlx5e_umr_dma_info, which used to contain only this array. The
flexibility of dma_info will allow to choose its size dynamically in a
following commit.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Instead of WARNing in runtime when TLS offload WQEs posted to ICOSQ are
over the hardware limit, check their size before enabling TLS RX
offload, and block the offload if the condition fails. It also allows to
drop a u16 field from struct mlx5e_icosq.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
To shorten and simplify code, let mlx5e_get_sw_max_sq_mpw_wqebbs accept
mdev and derive max SQ WQEBBs from it. Also rename the function to a
more generic name mlx5e_get_max_sq_aligned_wqebbs, because the following
patches will use it in non-MPWQE contexts.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
No need to keep max_sq_wqebbs in mlx5e_txqsq and mlx5e_xdpsq, as it's
only used when allocating the queues. Removing an extra field reduces
the struct size.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The return value of mlx5e_get_max_sq_wqebbs is clamped down to
MLX5_SEND_WQE_MAX_WQEBBS = 16, which fits into u8. This commit changes
the return type of this function to u8 for stricter type safety.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Expose mlx5e_create_mkey function, for future patches in the
macsec series to use.
The above function creates a memory key which describes a
region in memory that can be later used by both HW and SW.
The counterpart destroy functionality is already exposed.
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Emeel Hakim <ehakim@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This patch adds support for Connect-X MACsec offload Tx SA commands:
add, update and delete.
In Connect-X MACsec, a Security Association (SA) is added or deleted
via allocating a HW context of an encryption/decryption key and
a HW context of a matching SA (MACsec object).
When new SA is added:
- Use a separate crypto key HW context.
- Create a separate MACsec context in HW to include the SA properties.
Introduce a new compilation flag MLX5_EN_MACSEC for it.
Follow-up patches will implement the Tx steering.
Signed-off-by: Lior Nahmanson <liorna@nvidia.com>
Reviewed-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Make flow steering files fs_tt_redirect.c/h independent of en.h
such that it goes through the flow steering API only.
Make error reports be via mlx5_core API instead of netdev_err API, this
to ensure a safe decoupling from en.h, and prevent redundant argument
passing.
Signed-off-by: Lama Kayal <lkayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Conflicts:
net/ax25/af_ax25.c
d7c4c9e075f8c ("ax25: fix incorrect dev_tracker usage")
d62607c3fe459 ("net: rename reference+tracking helpers")
drivers/net/netdevsim/fib.c
180a6a3ee60a ("netdevsim: fib: Fix reference count leak on route deletion failure")
012ec02ae441 ("netdevsim: convert driver to use unlocked devlink API during init/fini")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Move the function declaration of mlx5e_init_l2_addr to en/fs.h, rename
to mlx5e_fs_init_l2_addr to align with the fs API functions naming
convention and let it take mlx5e_flow_steering as arguments while keeping
implementation at en_fs.c file. This helps maintain a clean driver code
and avoids unnecessary dependencies.
Signed-off-by: Lama Kayal <lkayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Make mlx5e_flow_steering member of mlx5e_priv a pointer.
Add dynamic allocation respectively.
Allocate fs for all profiles when initializing profile,
symmetrically deallocate at profile cleanup.
Signed-off-by: Lama Kayal <lkayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Introduce allocation and de-allocation functions for both flow steering
VLAN and TC as part of fs API.
Add allocations of VLAN and TC as nic profile feature, such that
fs_init() will allocate both VLAN and TC only if they're featured in
the profile. VLAN and TC are relevant for nic_profile only.
Signed-off-by: Lama Kayal <lkayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Before commit 76c31e5f7585 ("net/mlx5e: Use FW limitation for max MPW
WQEBBs"), the maximum size of MPWQE in WQEBBs was hardcoded as a driver
constant. That commit started using the firmware capability that can
further limit the size, however, it unintentionally changed a few
things:
1. The calculation of MLX5E_MAX_KLM_PER_WQE used the size in DS, which
was replaced by the size in WQEBBs, making the resulting value 4 times
smaller.
2. MLX5E_TX_MPW_MAX_WQEBBS used to be aligned to the cache line size
(either 64 or 128 bytes, i.e. 1 or 2 WQEBBs), but it's no longer the
case if the firmware capability is smaller than the driver maximum.
Fix both issues by using the correct units for MLX5E_MAX_KLM_PER_WQE and
by aligning mlx5e_get_sw_max_sq_mpw_wqebbs after taking the minimum.
Besides fixing the arithmetics in calculation of MLX5E_MAX_KLM_PER_WQE,
also use appropriate constants: `size of BSF * num of DS per WQEBB *
number of WQEBBs` (the calculation before the blamed commit) doesn't
make much sense to calculate the WQE size in bytes, so just use `size of
WQEBB * number of WQEBBs`.
While at it, replace the types that hold the number of WQEBBs by u8.
These values don't exceed 16, and it allows to fill holes in two
structs.
Fixes: 76c31e5f7585 ("net/mlx5e: Use FW limitation for max MPW WQEBBs")
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
MLX5E_MAX_RQ_NUM_MTTS should be the maximum value, so that
MLX5_MTT_OCTW(MLX5E_MAX_RQ_NUM_MTTS) fits into u16. The current value of
1 << 17 results in MLX5_MTT_OCTW(1 << 17) = 1 << 16, which doesn't fit
into u16. This commit replaces it with the maximum value that still
fits u16.
Fixes: 73281b78a37a ("net/mlx5e: Derive Striding RQ size from MTU")
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Move structure mlx5e_htb from the main driver include file "en.h" to be
hidden in qos.c where the qos functionality is implemented, forward
declare it for the rest of the driver and allocate it dynamically upon
user demand only.
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Moshe Tal <moshet@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com>
|
|
Preparation for dynamic allocation of the HTB struct.
The statistics should be preserved even when the struct is de-allocated.
Signed-off-by: Moshe Tal <moshet@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
HTB id fields are needed for selecting queue. Moving them to the
selq_params struct will simplify synchronization between control flow
and mlx5e_select_queues and will keep the IDs in the hot cacheline of
mlx5e_selq_params.
Replace mlx5e_selq_prepare() with separate functions that change subsets
of parameters, while keeping the rest.
This also will be useful to hide mlx5e_htb structure from the rest of the
driver in a later patch in this series.
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Moshe Tal <moshet@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com>
|
|
Keep mqprio_rl data to params and restore the configuration in case of
devlink reload.
Change the location of mqprio_rl resources cleanup so it will be done
also in reload flow.
Also, remove the rl pointer from the params, since this is dynamic object
and saved to priv.
Signed-off-by: Moshe Tal <moshet@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
HW-GRO (SHAMPO) packet merger scheme implies header-data split in the
driver, report it through the ethtool interface.
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
When the driver activates the channels, it assumes NAPI isn't running
yet. mlx5e_activate_rq posts a NOP WQE to ICOSQ to trigger a hardware
interrupt and start NAPI, which will run mlx5e_alloc_rx_mpwqe and post
UMR WQEs to ICOSQ to be able to receive packets with striding RQ.
Unfortunately, a race condition is possible if NAPI is triggered by
something else (for example, TX) at a bad timing, before
mlx5e_activate_rq finishes. In this case, mlx5e_alloc_rx_mpwqe may post
UMR WQEs to ICOSQ, and with the bad timing, the wqe_info of the first
UMR may be overwritten by the wqe_info of the NOP posted by
mlx5e_activate_rq.
The consequence is that icosq->db.wqe_info[0].num_wqebbs will be changed
from MLX5E_UMR_WQEBBS to 1, disrupting the integrity of the array-based
linked list in wqe_info[]. mlx5e_poll_ico_cq will hang in an infinite
loop after processing wqe_info[0], because after the corruption, the
next item to be processed will be wqe_info[1], which is filled with
zeros, and `sqcc += wi->num_wqebbs` will never move further.
This commit fixes this race condition by using async_icosq to post the
NOP and trigger the interrupt. async_icosq is always protected with a
spinlock, eliminating the race condition.
Fixes: bc77b240b3c5 ("net/mlx5e: Add fragmented memory support for RX multi packet WQE")
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reported-by: Karsten Nielsen <karsten@foo-bar.dk>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Correct the calculation of maximum channels of rep to better utilize
the hardware resources and allow a larger scale of reps.
This will allow creation of all virtual ports configured.
Fixes: 473baf2e9e8c ("net/mlx5e: Allow profile-specific limitation on max num of channels")
Signed-off-by: Moshe Tal <moshet@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
This commit removes the redundant check and removes the unused cqe parameter
of skb_from_cqe handlers.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Mellanox INNOVA TLS cards are EOL in May, 2018 [1]. As such, the code
is unmaintained, untested and not in-use by any upstream/distro oriented
customers. In order to reduce code complexity, drop the kernel code.
[1] https://network.nvidia.com/related-docs/eol/LCR-000286.pdf
Link: https://lore.kernel.org/r/b88add368def721ea9d054cb69def72d9e3f67aa.1649073691.git.leonro@nvidia.com
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
xmit_xdp_frame is extended to support sending fragmented XDP frames. The
next commit will start using this functionality.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
When MPWQE is disabled, mlx5e_open_xdpsq() prefills the common fields of
WQEs in the XDP SQ to save time when sending packets.
mlx5e_xmit_xdp_frame() runs on the prefilled fields, however, sending
multi buffer XDP frames would require changing some of these fields on a
per-packet basis. Besides that, mlx5e_xmit_xdp_frame() will be used as a
fallback to send multi buffer XDP frames when MPWQE is enabled (MPWQE
can only handle linear packets).
In order to prepare for XDP multi buffer support, this commit introduces
a mode for mlx5e_xmit_xdp_frame() that fills all the fields itself.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
The implementations of xmit_xdp_frame get the xdpi parameter of type
struct mlx5e_xdp_info for the sole purpose of calling
mlx5e_xdpi_fifo_push() on success.
This commit moves this call outside of xmit_xdp_frame, shifting this
responsibility to the caller. It will allow more fine-grained handling
of XDP info for cases when an xdp_frame is fragmented.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Use page_pool_set_dma_addr() to store the DMA address of a page inside
struct page, in order to avoid passing struct mlx5e_dma_info to XDP
handlers. Previously, struct mlx5e_dma_info was used to pass both the
DMA address and the page, and it worked well for the single-fragment
case.
When XDP multi buffer is in use, and a fragmented xdp_frame has to be
transmitted, the driver needs to know the DMA addresses of fragments,
however, the array of fragments in struct skb_shared_info doesn't
contain them. In order to pass the DMA addresses, the driver puts them
into struct page itself, which is accessible from the array of fragments
in struct skb_shared_info. The existing XDP handlers are modified to
remove the dependency on struct mlx5e_dma_info.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
In RQs of type multi-packet WQE (Striding RQ), each WQE is relatively
large (typically 256KB) but their number is relatively small (8 in
default).
Re-mapping the descriptors' buffers before re-posting them is done via
UMR (User-Mode Memory Registration) operations.
On the one hand, posting UMR WQEs in bulks reduces communication overhead
with the HW and better utilizes its processing units.
On the other hand, delaying the WQE repost operations for a small RQ
(say, of 4 WQEs) might drastically hit its performance, causing packet
drops due to no receive buffer, for high or bursty incoming packets rate.
Here we restrict the bulk size for too small RQs. Effectively, with the current
constants, RQ of size 4 (minimum allowed) would have no bulking, while larger
RQs will continue working with bulks of 2.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Start using the select queue parameters introduced in the previous
commit to have proper synchronization with changing the configuration
(such as number of channels and queues). It ensures that the state that
mlx5e_select_queue() sees is always consistent and stays the same while
the function is running. Also it allows mlx5e_select_queue to stop using
data structures that weren't synchronized properly: txq2sq,
channel_tc2realtxq, port_ptp_tc2realtxq. The last two are removed
completely, as they were used only in mlx5e_select_queue.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
ndo_select_queue can be called at any time, and there is no way to stop
the kernel from calling it to synchronize with configuration changes
(real_num_tx_queues, num_tc). This commit introduces an internal way in
mlx5e to sync mlx5e_select_queue() with these changes. The configuration
needed by this function is stored in a struct mlx5e_selq_params, which
is modified and accessed in an atomic way using RCU methods. The whole
ndo_select_queue is called under an RCU lock, providing the necessary
guarantees.
The parameters stored in the new struct mlx5e_selq_params should only be
used from inside mlx5e_select_queue. It's the minimal set of parameters
needed for mlx5e_select_queue to do its job efficiently, derived from
parameters stored elsewhere. That means that when the configuration
change, mlx5e_selq_params may need to be updated. In such cases, the
mlx5e_selq_prepare/mlx5e_selq_apply API should be used.
struct mlx5e_selq contains two slots for the params: active and standby.
mlx5e_selq_prepare updates the standby slot, and mlx5e_selq_apply swaps
the slots in a safe atomic way using the RCU API. It integrates well
with the open/activate stages of the configuration change flow.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Calculate maximal count of MPW WQEBBs on SQ's creation and store it
there. Remove MLX5E_TX_MPW_MAX_NUM_DS and MLX5E_TX_MPW_MAX_WQEBBS.
Update mlx5e_tx_mpwqe_is_full() and mlx5e_xdp_mpqwe_is_full() .
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Prior to this patch the maximal value for max WQEBBs (WQE Basic Blocks,
where WQE is a Work Queue Element) on the TX side was assumed to be 16
(fixed value). All firmware versions till today comply to this. In order
to be more flexible and resilient, read from FW the corresponding:
max_wqe_sz_sq. This value describes the maximum WQE size given in bytes,
thus max WQEBBs is given by the division in WQEBB's byte size. The
driver uses the top between 16 and the division result. This ensures
synchronization between driver and firmware and avoids unexpected
behavior. Store this value on the different SQs (Send Queues) for easy
access.
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Remove tstamp pointer in mlx5e_txqsq as it's no longer used after
commit 7c39afb394c7 ("net/mlx5: PTP code migration to driver core section").
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Aya Levin <ayal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally writing across neighboring fields.
Use flexible arrays instead of zero-element arrays (which look like they
are always overflowing) and split the cross-field memcpy() into two halves
that can be appropriately bounds-checked by the compiler.
We were doing:
#define ETH_HLEN 14
#define VLAN_HLEN 4
...
#define MLX5E_XDP_MIN_INLINE (ETH_HLEN + VLAN_HLEN)
...
struct mlx5e_tx_wqe *wqe = mlx5_wq_cyc_get_wqe(wq, pi);
...
struct mlx5_wqe_eth_seg *eseg = &wqe->eth;
struct mlx5_wqe_data_seg *dseg = wqe->data;
...
memcpy(eseg->inline_hdr.start, xdptxd->data, MLX5E_XDP_MIN_INLINE);
target is wqe->eth.inline_hdr.start (which the compiler sees as being
2 bytes in size), but copying 18, intending to write across start
(really vlan_tci, 2 bytes). The remaining 16 bytes get written into
wqe->data[0], covering byte_count (4 bytes), lkey (4 bytes), and addr
(8 bytes).
struct mlx5e_tx_wqe {
struct mlx5_wqe_ctrl_seg ctrl; /* 0 16 */
struct mlx5_wqe_eth_seg eth; /* 16 16 */
struct mlx5_wqe_data_seg data[]; /* 32 0 */
/* size: 32, cachelines: 1, members: 3 */
/* last cacheline: 32 bytes */
};
struct mlx5_wqe_eth_seg {
u8 swp_outer_l4_offset; /* 0 1 */
u8 swp_outer_l3_offset; /* 1 1 */
u8 swp_inner_l4_offset; /* 2 1 */
u8 swp_inner_l3_offset; /* 3 1 */
u8 cs_flags; /* 4 1 */
u8 swp_flags; /* 5 1 */
__be16 mss; /* 6 2 */
__be32 flow_table_metadata; /* 8 4 */
union {
struct {
__be16 sz; /* 12 2 */
u8 start[2]; /* 14 2 */
} inline_hdr; /* 12 4 */
struct {
__be16 type; /* 12 2 */
__be16 vlan_tci; /* 14 2 */
} insert; /* 12 4 */
__be32 trailer; /* 12 4 */
}; /* 12 4 */
/* size: 16, cachelines: 1, members: 9 */
/* last cacheline: 16 bytes */
};
struct mlx5_wqe_data_seg {
__be32 byte_count; /* 0 4 */
__be32 lkey; /* 4 4 */
__be64 addr; /* 8 8 */
/* size: 16, cachelines: 1, members: 3 */
/* last cacheline: 16 bytes */
};
So, split the memcpy() so the compiler can reason about the buffer
sizes.
"pahole" shows no size nor member offset changes to struct mlx5e_tx_wqe
nor struct mlx5e_umr_wqe. "objdump -d" shows no meaningful object
code changes (i.e. only source line number induced differences and
optimizations).
Fixes: b5503b994ed5 ("net/mlx5e: XDP TX forwarding support")
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|