Age | Commit message (Collapse) | Author |
|
Enhance the memory scheme of the legacy RQ, such that
only order-0 pages are used.
Whenever possible, prefer using a linear SKB, and build it
wrapping the WQE buffer.
Otherwise (for example, jumbo frames on x86), use non-linear SKB,
with as many frags as needed. In this case, multiple WQE
scatter entries are used, up to a maximum of 4 frags and 10KB of MTU.
This implied to remove support of HW LRO in legacy RQ, as it would
require large number of page allocations and scatter entries per WQE
on archs with PAGE_SIZE = 4KB, yielding bad performance.
In earlier patches, we guaranteed that all completions are in-order,
and that we use a cyclic WQ.
This creates an oppurtunity for a performance optimization:
The mapping between a "struct mlx5e_dma_info", and the
WQEs (struct mlx5e_wqe_frag_info) pointing to it, is constant
across different cycles of a WQ. This allows initializing
the mapping in the time of RQ creation, and not handle it
in datapath.
A struct mlx5e_dma_info that is shared between different WQEs
is allocated by the first WQE, and freed by the last one.
This implies an important requirement: WQEs that share the same
struct mlx5e_dma_info must be posted within the same NAPI.
Otherwise, upon completion, struct mlx5e_wqe_frag_info would mistakenly
point to the new struct mlx5e_dma_info, not the one that was posted
(and the HW wrote to).
This bulking requirement is actually good also for performance reasons,
hence we extend the bulk beyong the minimal requirement above.
With this memory scheme, the RQs memory footprint is reduce by a
factor of 2 on x86, and by a factor of 32 on PowerPC.
Same factors apply for the number of pages in a GRO session.
Performance tests:
ConnectX-4, single core, single RX ring, default MTU.
x86:
CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
Packet rate (early drop in TC): no degradation
TCP streams: ~5% improvement
PowerPC:
CPU: POWER8 (raw), altivec supported
Packet rate (early drop in TC): 20% gain
TCP streams: 25% gain
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Now that LRO is not supported for Legacy RQ, there is no source of
out-of-order completions in the WQ, and we can use a cyclic one.
This has multiple advantages:
- reduces the WQE size (smaller PCI transactions).
- lower overhead in datapath (no handling of 'next' pointers).
- no reserved WQE for the WQ head (was need in linked-list).
- allows using a constant map between frag and dma_info struct, in downstream patch.
Performance tests:
ConnectX-4, single core, single RX ring.
Major gain in packet rate of single ring XDP drop.
Bottleneck is shifted form HW (at 16Mpps) to SW (at 20Mpps).
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Replace the common RQ WQ object with two separate ones for the
different RQ types.
This is in preparation for switching to using a cyclic WQ type
in Legacy RQ.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Get the logic of copying the packet header into the SKB linear part
into a generic function. Function does copy length alignment
and dma buffer sync.
It is currently called only within the MPWQE flow.
In a downstream patch, it will be called within the legacy RQ flow
as well.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Rename it and pass truesize as an extra argument, as it will be used also
in Legacy RQ in a downstream patch.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Make name more generic by dropping MPWRQ from it, as it will be
used also in Legacy RQ in a downstream patch.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5e-updates-2018-05-25
This series includes updates for mlx5e netdev driver.
1) Allowr flow based VF vport mirroring under sriov switchdev scheme,
added support for offloading the TC mirred mirror sub-action, from
Chris Mi.
=================
From: Or Gerlitz <ogerlitz@mellanox.com>
The user will typically set the actions order such that the mirror
port (mirror VF) sees packets as the original port (VF under
mirroring) sent them or as it will receive them. In the general case,
it means that packets are potentially sent to the mirror port before
or after some actions were applied on them.
To properly do that, we follow on the exact action order as set for
the flow and make sure this will also be the case when we program the
HW offload.
If all the actions should apply before forwarding to the mirror and dest port,
mirroring is just multicasting to the two vports. Otherwise, we split
the TC flow to two HW rules, where the 1st applies only the actions
needed up to the mirror (if there are such) and the 2nd the rest of
the actions plus the forwarding to the dest vport.
=================
2) Move to order-0 only allocations (using fragmented work queues) for all
work queues used by the driver, RX and TX descriptor rings
(RQs, SQs and Completion Queues (CQs)), from Tariq Toukan.
3) Avoid resetting netdevice statistics on netdevice
state changes, from Eran Ben Elisha.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Lots of easy overlapping changes in the confict
resolutions here.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Move all RQ, SQ and channel counters from the channel objects into the
priv structure. With this change, counters will not be reset upon
channel configuration changes.
Channel's statistics for SQs which are associated with TCs higher than
zero will be presented in ethtool -S, only for SQs which were opened at
least once since the module was loaded (regardless of their open/close
current status). This is done in order to decrease the total amount of
statistics presented and calculated for the common out of box use (no
QoS).
mlx5e_channel_stats is a compound of CH,RQ,SQs stats in order to
create locality for the NAPI when handling TX and RX of the same
channel.
Align the new statistics struct per ring to avoid several channels
update to the same cache line at the same time.
Packet rate was tested, no degradation sensed.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
CC: Qing Huang <qing.huang@oracle.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Complete the transition of all WQ types to use fragmented
order-0 coherent memory instead of high-order allocations.
CQ-WQ already uses order-0.
Here we do the same for cyclic and linked-list WQs.
This allows the driver to load cleanly on systems with a highly
fragmented coherent memory.
Performance tests:
ConnectX-5 100Gbps, CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
Packet rate of 64B packets, single transmit ring, size 8K.
No degradation is sensed.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
We fill SQ edge with NOPs to avoid WQEs wrap.
Here, instead of doing that in advance for the maximum possible
WQE size, we do it on-demand using the actual WQE size.
We re-order some parts in mlx5e_sq_xmit to finish the calculation
of WQE size (ds_cnt) before doing any writes to the WQE buffer.
When SQ work queue is fragmented (introduced in an downstream patch),
dealing with WQE wraps becomes more frequent. This change would drastically
reduce the overhead in this case.
Performance tests:
ConnectX-5 100Gbps, CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
Packet rate of 64B packets, single transmit ring, size 8K.
Before: 14.9 Mpps
After: 15.8 Mpps
Improvement of 6%.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Use the WQ API to get the WQ size, and to map a counter
into a WQ entry index.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
When RXFCS feature is enabled, the HW do not strip the FCS data,
however it is not present in the checksum calculated by the HW.
Fix that by manually calculating the FCS checksum and adding it to the SKB
checksum field.
Add helper function to find the FCS data for all SKB forms (linear,
one fragment or more).
Fixes: 102722fc6832 ("net/mlx5e: Add support for RXFCS feature flag")
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
MLX5E_TEST_BIT macro is the same as the already existent test_bit,
remove it and replace all usages.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Function returns boolean values, use bool instead of int.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Range of LRO number of segments fits in u8.
Also, bring initialization and declaration together to
save code.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Changing API xdp_return_frame() to take struct xdp_frame as argument,
seems like a natural choice. But there are some subtle performance
details here that needs extra care, which is a deliberate choice.
When de-referencing xdp_frame on a remote CPU during DMA-TX
completion, result in the cache-line is change to "Shared"
state. Later when the page is reused for RX, then this xdp_frame
cache-line is written, which change the state to "Modified".
This situation already happens (naturally) for, virtio_net, tun and
cpumap as the xdp_frame pointer is the queued object. In tun and
cpumap, the ptr_ring is used for efficiently transferring cache-lines
(with pointers) between CPUs. Thus, the only option is to
de-referencing xdp_frame.
It is only the ixgbe driver that had an optimization, in which it can
avoid doing the de-reference of xdp_frame. The driver already have
TX-ring queue, which (in case of remote DMA-TX completion) have to be
transferred between CPUs anyhow. In this data area, we stored a
struct xdp_mem_info and a data pointer, which allowed us to avoid
de-referencing xdp_frame.
To compensate for this, a prefetchw is used for telling the cache
coherency protocol about our access pattern. My benchmarks show that
this prefetchw is enough to compensate the ixgbe driver.
V7: Adjust for commit d9314c474d4f ("i40e: add support for XDP_REDIRECT")
V8: Adjust for commit bd658dda4237 ("net/mlx5e: Separate dma base address
and offset in dma_sync call")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This patch shows how it is possible to have both the driver local page
cache, which uses elevated refcnt for "catching"/avoiding SKB
put_page returns the page through the page allocator. And at the
same time, have pages getting returned to the page_pool from
ndp_xdp_xmit DMA completion.
The performance improvement for XDP_REDIRECT in this patch is really
good. Especially considering that (currently) the xdp_return_frame
API and page_pool_put_page() does per frame operations of both
rhashtable ID-lookup and locked return into (page_pool) ptr_ring.
(It is the plan to remove these per frame operation in a followup
patchset).
The benchmark performed was RX on mlx5 and XDP_REDIRECT out ixgbe,
with xdp_redirect_map (using devmap) . And the target/maximum
capability of ixgbe is 13Mpps (on this HW setup).
Before this patch for mlx5, XDP redirected frames were returned via
the page allocator. The single flow performance was 6Mpps, and if I
started two flows the collective performance drop to 4Mpps, because we
hit the page allocator lock (further negative scaling occurs).
Two test scenarios need to be covered, for xdp_return_frame API, which
is DMA-TX completion running on same-CPU or cross-CPU free/return.
Results were same-CPU=10Mpps, and cross-CPU=12Mpps. This is very
close to our 13Mpps max target.
The reason max target isn't reached in cross-CPU test, is likely due
to RX-ring DMA unmap/map overhead (which doesn't occur in ixgbe to
ixgbe testing). It is also planned to remove this unnecessary DMA
unmap in a later patchset
V2: Adjustments requested by Tariq
- Changed page_pool_create return codes not return NULL, only
ERR_PTR, as this simplifies err handling in drivers.
- Save a branch in mlx5e_page_release
- Correct page_pool size calc for MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ
V5: Updated patch desc
V8: Adjust for b0cedc844c00 ("net/mlx5e: Remove rq_headroom field from params")
V9:
- Adjust for 121e89275471 ("net/mlx5e: Refactor RQ XDP_TX indication")
- Adjust for 73281b78a37a ("net/mlx5e: Derive Striding RQ size from MTU")
- Correct handling if page_pool_create fail for MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ
V10: Req from Tariq
- Change pool_size calc for MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This implements basic XDP redirect support in mlx5 driver.
Notice that the ndo_xdp_xmit() is NOT implemented, because that API
need some changes that this patchset is working towards.
The main purpose of this patch is have different drivers doing
XDP_REDIRECT to show how different memory models behave in a cross
driver world.
Update(pre-RFCv2 Tariq): Need to DMA unmap page before xdp_do_redirect,
as the return API does not exist yet to to keep this mapped.
Update(pre-RFCv3 Saeed): Don't mix XDP_TX and XDP_REDIRECT flushing,
introduce xdpsq.db.redirect_flush boolian.
V9: Adjust for commit 121e89275471 ("net/mlx5e: Refactor RQ XDP_TX indication")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Upon a new UMR post, check if the WQE buffer contains
a previous UMR WQE. If so, modify the dynamic fields
instead of a whole WQE overwrite. This saves a memcpy.
In current setting, after 2 WQ cycles (12 UMR posts),
this will always be the case.
No degradation sensed.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
All UMR WQEs of an RQ share many common fields. We use
pre-initialized structures to save calculations in datapath.
One field (xlt_offset) was the only reason we saved a pre-initialized
copy per WQE index.
Here we remove its initialization (move its calculation to datapath),
and reduce the number of copies to one-per-RQ.
A very small datapath calculation is added, it occurs once per a MPWQE
(i.e. once every 256KB), but reduces memory consumption and gives
better cache utilization.
Performance testing:
Tested packet rate, no degradation sensed.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
When many packets reside on the same page, the bulking of
page_ref modifications reduces the total number of atomic
operations executed.
Besides the necessary 2 operations on page alloc/free, we
have the following extra ops per page:
- one on WQE allocation (bump refcnt to maximum possible),
- zero ops for SKBs,
- one on WQE free,
a constant of two operations in total, no matter how many
packets/SKBs actually populate the page.
Without this bulking, we have:
- no ops on WQE allocation or free,
- one op per SKB,
Comparing the two methods when PAGE_SIZE is 4K:
- As mentioned above, bulking method always executes 2 operations,
not more, but not less.
- In the default MTU configuration (1500, stride size is 2K),
the non-bulking method execute 2 ops as well.
- For larger MTUs with stride size of 4K, non-bulking method
executes only a single op.
- For XDP (stride size of 4K, no SKBs), non-bulking method
executes no ops at all!
Hence, to optimize the flows with linear SKB and XDP over Striding RQ,
we here remove the page_ref bulking method.
Performance testing:
ConnectX-5, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz.
Single core packet rate (64 bytes).
Early drop in TC: no degradation.
XDP_DROP:
before: 14,270,188 pps
after: 20,503,603 pps, 43% improvement.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Add XDP support over Striding RQ.
Now that linear SKB is supported over Striding RQ,
we can support XDP by setting stride size to PAGE_SIZE
and headroom to XDP_PACKET_HEADROOM.
Upon a MPWQE free, do not release pages that are being
XDP xmit, they will be released upon completions.
Striding RQ is capable of a higher packet-rate than
conventional RQ.
A performance gain is expected for all cases that had
a HW packet-rate bottleneck. This is the case whenever
using many flows that distribute to many cores.
Performance testing:
ConnectX-5, 24 rings, default MTU.
CQE compression ON (to reduce completions BW in PCI).
XDP_DROP packet rate:
--------------------------------------------------
| pkt size | XDP rate | 100GbE linerate | pct% |
--------------------------------------------------
| 64byte | 126.2 Mpps | 148.0 Mpps | 85% |
| 128byte | 80.0 Mpps | 84.8 Mpps | 94% |
| 256byte | 42.7 Mpps | 42.7 Mpps | 100% |
| 512byte | 23.4 Mpps | 23.4 Mpps | 100% |
--------------------------------------------------
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Make the xdp_xmit indication available for Striding RQ
by taking it out of the type-specific union.
This refactor is a preparation for a downstream patch that
adds XDP support over Striding RQ.
In addition, use a bitmap instead of a boolean for possible
future flags.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Current Striding RQ HW feature utilizes the RX buffers so that
there is no wasted room between the strides. This maximises
the memory utilization.
This prevents the use of build_skb() (which requires headroom
and tailroom), and demands to memcpy the packets headers into
the skb linear part.
In this patch, whenever a set of conditions holds, we apply
an RQ configuration that allows combining the use of linear SKB
on top of a Striding RQ.
To use build_skb() with Striding RQ, the following must hold:
1. packet does not cross a page boundary.
2. there is enough headroom and tailroom surrounding the packet.
We can satisfy 1 and 2 by configuring:
stride size = MTU + headroom + tailoom.
This is possible only when:
a. (MTU - headroom - tailoom) does not exceed PAGE_SIZE.
b. HW LRO is turned off.
Using linear SKB has many advantages:
- Saves a memcpy of the headers.
- No page-boundary checks in datapath.
- No filler CQEs.
- Significantly smaller CQ.
- SKB data continuously resides in linear part, and not split to
small amount (linear part) and large amount (fragment).
This saves datapath cycles in driver and improves utilization
of SKB fragments in GRO.
- The fragments of a resulting GRO SKB follow the IP forwarding
assumption of equal-size fragments.
Some implementation details:
HW writes the packets to the beginning of a stride,
i.e. does not keep headroom. To overcome this we make sure we can
extend backwards and use the last bytes of stride i-1.
Extra care is needed for stride 0 as it has no preceding stride.
We make sure headroom bytes are available by shifting the buffer
pointer passed to HW by headroom bytes.
This configuration now becomes default, whenever capable.
Of course, this implies turning LRO off.
Performance testing:
ConnectX-5, single core, single RX ring, default MTU.
UDP packet rate, early drop in TC layer:
--------------------------------------------
| pkt size | before | after | ratio |
--------------------------------------------
| 1500byte | 4.65 Mpps | 5.96 Mpps | 1.28x |
| 500byte | 5.23 Mpps | 5.97 Mpps | 1.14x |
| 64byte | 5.94 Mpps | 5.96 Mpps | 1.00x |
--------------------------------------------
TCP streams: ~20% gain
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
When modifying the page mapping of a HW memory region
(via a UMR post), post the new values inlined in WQE,
instead of using a data pointer.
This is a micro-optimization, inline UMR WQEs of different
rings scale better in HW.
In addition, this obsoletes a few control flows and helps
delete ~50 LOC.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Do not busy-wait a pending UMR completion. Under high HW load,
busy-waiting a delayed completion would fully utilize the CPU core
and mistakenly indicate a SW bottleneck.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Gets the process of a UMR WQE post in one function,
in preparation for a downstream patch that inlines
the WQE data.
No functional change here.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Knowing the MTU is required for RQ creation flow.
By our design, channels creation flow is totally isolated
from priv/netdev, and can be completed with access to
channels params and mdev.
Adding the MTU to the channels params helps preserving that.
In addition, we save it in RQ to make its access faster in
datapath checks.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
In copying skb header to skb->data, replace the call to
skb_copy_to_linear_data_offset() with a zero offset with
the call to the no-offset function skb_copy_to_linear_data().
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Pass the base dma address and offset to dma_sync_single_range_for_cpu(),
instead of doing the pre-calculation.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
|
|
When receiving an LRO packet, the checksum field is set by the hardware
to the checksum of the first coalesced packet. Obviously, this checksum
is not valid for the merged LRO packet and should be fixed. We can use
the CQE checksum which covers the checksum of the entire merged packet
TCP payload to help us calculate the checksum incrementally.
Tested by sending IPv4/6 traffic with LRO enabled, RX checksum disabled
and watching nstat checksum error counters (in addition to the obvious
bandwidth drop caused by checksum errors).
This bug is usually "hidden" since LRO packets would go through the
CHECKSUM_UNNECESSARY flow which does not validate the packet checksum.
It's important to note that previous to this patch, LRO packets provided
with CHECKSUM_UNNECESSARY are indeed packets with a correct validated
checksum (even though the checksum inside the TCP header is incorrect),
since the hardware LRO aggregation is terminated upon receiving a packet
with bad checksum.
Fixes: e586b3b0baee ("net/mlx5: Ethernet Datapath files")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
The current implementation of create CQ requires contiguous
memory, such requirement is problematic once the memory is
fragmented or the system is low in memory, it causes for
failures in dma_zalloc_coherent().
This patch implements new scheme of fragmented CQ to overcome
this issue by introducing new type: 'struct mlx5_frag_buf_ctrl'
to allocate fragmented buffers, rather than contiguous ones.
Base the Completion Queues (CQs) on this new fragmented buffer.
It fixes following crashes:
kworker/29:0: page allocation failure: order:6, mode:0x80d0
CPU: 29 PID: 8374 Comm: kworker/29:0 Tainted: G OE 3.10.0
Workqueue: ib_cm cm_work_handler [ib_cm]
Call Trace:
[<>] dump_stack+0x19/0x1b
[<>] warn_alloc_failed+0x110/0x180
[<>] __alloc_pages_slowpath+0x6b7/0x725
[<>] __alloc_pages_nodemask+0x405/0x420
[<>] dma_generic_alloc_coherent+0x8f/0x140
[<>] x86_swiotlb_alloc_coherent+0x21/0x50
[<>] mlx5_dma_zalloc_coherent_node+0xad/0x110 [mlx5_core]
[<>] ? mlx5_db_alloc_node+0x69/0x1b0 [mlx5_core]
[<>] mlx5_buf_alloc_node+0x3e/0xa0 [mlx5_core]
[<>] mlx5_buf_alloc+0x14/0x20 [mlx5_core]
[<>] create_cq_kernel+0x90/0x1f0 [mlx5_ib]
[<>] mlx5_ib_create_cq+0x3b0/0x4e0 [mlx5_ib]
Signed-off-by: Yonatan Cohen <yonatanc@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Most of the packets return true for is_last_ethertype_ip, surround it
with likely compiler hint.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
The current implementation takes the child timestamp object from
the parent since the rq in mlx5i_complete_rx_cqe belongs to the parent.
This change fixes the issue by taking the correct timestamp.
Fixes: 7e7f4780c340 ("net/mlx5e: IPoIB, Use hash-table to map between QPN to child netdev")
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Use the more appropriate netdev_WARN_ONCE instead of WARN_ONCE macro.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Reviewed-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The mlx5 driver have a special drop-RQ queue (one per interface) that
simply drops all incoming traffic. It helps driver keep other HW
objects (flow steering) alive upon down/up operations. It is
temporarily pointed by flow steering objects during the interface
setup, and when interface is down. It lacks many fields that are set
in a regular RQ (for example its state is never switched to
MLX5_RQC_STATE_RDY). (Thanks to Tariq Toukan for explanation).
The XDP RX-queue info for this drop-RQ marked as unused, which
allow us to use the same takedown/free code path as other RX-queues.
Driver hook points for xdp_rxq_info:
* reg : mlx5e_alloc_rq()
* unused: mlx5e_alloc_drop_rq()
* unreg : mlx5e_free_rq()
Tested on actual hardware with samples/bpf program
Cc: Saeed Mahameed <saeedm@mellanox.com>
Cc: Matan Barak <matanb@mellanox.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
|
|
Currently, when dma mapping fails, put_page is called,
but the page is not set to null. Later, in the page_reuse treatment in
mlx5e_free_rx_descs(), mlx5e_page_release() is called for the second time,
improperly doing dma_unmap (for a non-mapped address) and an extra put_page.
Prevent this by nullifying the page pointer when dma_map fails.
Fixes: accd58833237 ("net/mlx5e: Introduce RX Page-Reuse")
Signed-off-by: Inbar Karmy <inbark@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Cc: kernel-team@fb.com
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
When the VLAN tag is present in the packet buffer (i.e VLAN stripping disabled, QinQ)
the driver will currently report CHECKSUM_UNNECESSARY.
Instead of using CHECKSUM_COMPLETE offload for packets with first
ethertype of IPv4/6, use it for packets with last ethertype of IPv4/6 to
cover the former cases as well.
The checksum field present in the CQE is calculated from the IP header
until the end of the packet. When the first ethertype is different than
IPv4/6 (for ex. 802.1Q VLAN) a checksum of the VLAN header/s should be
added. The small header/s checksum calculation will allow us to use
CHECKSUM_COMPLETE instead of CHECKSUM_UNNECESSARY.
Testing bandwidth of one and 8 TCP streams to a single RQ,
LRO and VLAN stripping offloads disabled:
CPU: Intel(R) Xeon(R) CPU E5-2660 v2 @ 2.20GHz
NIC: Mellanox Technologies MT27710 Family [ConnectX-4 Lx]
Before:
+--------------+--------------------+---------------------+----------------------+
| Traffic type | 1 Stream BW [Mbps] | 8 Streams BW [Mbps] | Checksum offload |
+--------------+--------------------+---------------------+----------------------+
| Untagged | 28,247.35 | 24,716.88 | CHECKSUM_COMPLETE |
| VLAN | 27,516.69 | 23,752.26 | CHECKSUM_UNNECESSARY |
| QinQ | 6,961.30 | 20,667.04 | CHECKSUM_UNNECESSARY |
+--------------+--------------------+---------------------+----------------------+
Now:
+--------------+--------------------+---------------------+-------------------+
| Traffic type | 1 Stream BW [Mbps] | 8 Streams BW [Mbps] | Checksum offload |
+--------------+--------------------+---------------------+-------------------+
| Untagged | 28,521.28 | 24,926.32 | CHECKSUM_COMPLETE |
| VLAN | 27,389.37 | 23,715.34 | CHECKSUM_COMPLETE |
| QinQ | 6,901.77 | 20,845.73 | CHECKSUM_COMPLETE |
+--------------+--------------------+---------------------+-------------------+
No performance degradation observed.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
The following counters are now exposed through ethtool -S:
rx[i]_removed_vlan_packets (per channel)
rx_removed_vlan_packets
tx[i]_added_vlan_packets (per channel)
tx_added_vlan_packets
rx_removed_vlan_packets: The number of packets that had their
outer VLAN header stripped to the CQE by the hardware.
tx_added_vlan_packets: The number of packets that had their
outer VLAN header inserted by the hardware.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux
Saeed Mahameed says:
====================
mlx5-updates-2017-10-11: IPoIB Multi Pkey support
This series provides the support for IPoIB Multi Pkey.
InfiniBand Pkeys are the equivalent of Ethernet vlans.
Currently IPoIB device driver supports only default Pkey and IPoIB Pkey child
interfaces are not supported with IPoIB offloads mode, this series will add
the support for that by allowing creating mlx5 multiple IPoIB netdevices with
a non-default Pkey.
mlx5 IPoIB Pkey child interface is smaller version of mlx5i IPoIB interfaces and shares
most of its resources with the parent IPoIB interface, namely RX steering and ring
queue resources.
The only mlx5 resources a child Pkey interface will be creating are the TX rings,
since they should be assigned to a specific Pkey.
mlx5i Pkey netdev is implemented via new mlx5e netdev profile implemented in
mlx5/core/ipoib/ipoib_vlan.c.
The series starts with a refactoring of mlx5e PTP and mlx5 clock implementation
to move the code to be part of mlx5 core rather than mlx5e netdevice, in order to
make mlx5 clock and PTP registration part of the core to be shared with mlx5e
master Ethernet netdev/IPoIB parent netdev and mlx5_ib in the near future.
Add the support for attaching multiple underlay QPs for the different Pkeys
in mlx5 core RX steering.
Add Pkey index to rdma_netdev to add the ability to set PKEY index to lower
IPoIB offload netdev.
Use hash-table to map between DQPN (Destination QP number) to child netdev
for the IPoIB parent netdev to forward RX packets to the corresponding
child Pkey netdev, since the RX rings are shared.
The reset of the series adds the ipoib child Pkey: mlx5e netdev profile,
netdev nods implementation and minimal set of ethtool callbacks.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This change is needed for PKEY support, since the RQs are shared
between the child interface and the parent. The parent is responsible
for NAPI and the precessing of RX completions. Using the dqpn in the
completion descriptor we set the corresponding child IPoIB netdevice
on the SKB.
The mapping between the dqpn and the netdevice is done using a HT,
each mlx5 IPoIB interface registers its mapping on creation.
Signed-off-by: Alex Vesker <valex@mellanox.com>
Reviewed-by: Erez Shitrit <erezsh@mellanox.com>
|
|
PTP code is moved to core section of mlx5 driver in order to share
it between ethernet and infiniband. This movement involves the following
changes:
- Change mlx5e_ prefix to be mlx5_
- Add clock structs to Core
- Add clock object to mlx5_core_dev
- Call Init/Uninit clock from core init/cleanup
- Rename mlx5e_tstamp to be mlx5_clock
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Eitan Rabin <rabin@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
Just simple overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Instead of calculating the offloads counters, count them explicitly.
The calculations done for these counters would result in bugs in some
cases, for example:
When running TCP traffic over a VXLAN tunnel with TSO enabled the following
counters would increase:
tx_csum_partial: 1,333,284
tx_csum_partial_inner: 29,286
tx4_csum_partial_inner: 384
tx7_csum_partial_inner: 8
tx9_csum_partial_inner: 34
tx10_csum_partial_inner: 26,807
tx11_csum_partial_inner: 287
tx12_csum_partial_inner: 27
tx16_csum_partial_inner: 6
tx25_csum_partial_inner: 1,733
Seems like tx_csum_partial increased out of nowhere.
The issue is in the following calculation in mlx5e_update_sw_counters:
s->tx_csum_partial = s->tx_packets - tx_offload_none - s->tx_csum_partial_inner;
While tx_packets increases by the number of GSO segments for each SKB,
tx_csum_partial_inner will only increase by one, resulting in wrong
tx_csum_partial counter.
Fixes: bfe6d8d1d433 ("net/mlx5e: Reorganize ethtool statistics")
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
This work enables generic transfer of metadata from XDP into skb. The
basic idea is that we can make use of the fact that the resulting skb
must be linear and already comes with a larger headroom for supporting
bpf_xdp_adjust_head(), which mangles xdp->data. Here, we base our work
on a similar principle and introduce a small helper bpf_xdp_adjust_meta()
for adjusting a new pointer called xdp->data_meta. Thus, the packet has
a flexible and programmable room for meta data, followed by the actual
packet data. struct xdp_buff is therefore laid out that we first point
to data_hard_start, then data_meta directly prepended to data followed
by data_end marking the end of packet. bpf_xdp_adjust_head() takes into
account whether we have meta data already prepended and if so, memmove()s
this along with the given offset provided there's enough room.
xdp->data_meta is optional and programs are not required to use it. The
rationale is that when we process the packet in XDP (e.g. as DoS filter),
we can push further meta data along with it for the XDP_PASS case, and
give the guarantee that a clsact ingress BPF program on the same device
can pick this up for further post-processing. Since we work with skb
there, we can also set skb->mark, skb->priority or other skb meta data
out of BPF, thus having this scratch space generic and programmable
allows for more flexibility than defining a direct 1:1 transfer of
potentially new XDP members into skb (it's also more efficient as we
don't need to initialize/handle each of such new members). The facility
also works together with GRO aggregation. The scratch space at the head
of the packet can be multiple of 4 byte up to 32 byte large. Drivers not
yet supporting xdp->data_meta can simply be set up with xdp->data_meta
as xdp->data + 1 as bpf_xdp_adjust_meta() will detect this and bail out,
such that the subsequent match against xdp->data for later access is
guaranteed to fail.
The verifier treats xdp->data_meta/xdp->data the same way as we treat
xdp->data/xdp->data_end pointer comparisons. The requirement for doing
the compare against xdp->data is that it hasn't been modified from it's
original address we got from ctx access. It may have a range marking
already from prior successful xdp->data/xdp->data_end pointer comparisons
though.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Avoid recycling an RX page if it moved to another NUMA node.
Add an ethtool counter to count such events.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|
|
As of current design, in each NAPI, only a single UMR WQE
completion could be available in the completion queue of the
the internal control operations (ICO) send queue, in addition
to nop operations that require no actions upon completion.
This renders the consume index obsolete, as the wqe_counter
field in CQE is sufficient.
This helps removing a memory barrier, and obsoletes the need
for tracking the num_wqebbs to update the consumer counter.
In addition, remove other unused fields in icosq struct:
pdev, dma_fifo_pc, and prev_cc.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
|