Age | Commit message (Collapse) | Author | Files | Lines |
|
bnxt has 1:1 mapping of page pools and NAPIs, so it's safe
to hoook them up together.
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Tested-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Check if patterns and arguments for modify header action
are supported and enable them accordingly.
Signed-off-by: Muhammad Sammar <muhammads@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Support the pattern/args-based MODIFY_HDR and TNL_L3_TO_L2 actions in dbg dump
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Set modify header action of size 1 directly on the STE for supporting
devices, thus reducing number of hops and cache misses.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Use the new accelerated action for decap L3 on RX side:
use the mechanism of pattern and argument same as in
modify-header action.
Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
If there is support for pattern/args, use the new accelerated modify
header action for modify header and decap L3 actions.
Otherwise fall back to the old modify-header implementation.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
While building the actions, add the pointer of the arguments for
accelerated modify list action into the action's attributes.
This will be used later on while building the specific STE
for this action.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Added new mechanism for handling arguments for modify-header action.
The new action "accelerated modify-header" asks for the arguments from
separated area from the pattern, this area accessed via general objects.
Handling of these object is done via the pool-manager struct.
When the new header patterns are supported, while loading the domain,
a few pools for argument creations will be created. The requests for
allocating/deallocating arg objects are done via the pool manager API.
Signed-off-by: Muhammad Sammar <muhammads@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
When allocating a QP we allocate an RQ and an SQ, the RQ is stored first
in memory and followed by the SQ.
This allocation is not physically continiuos - it may span across different
physical pages. SW Steering code always writes in pairs: 1BB write + 1BB read,
or 2 continuous BBs of GTA WQE.
This lead to an issue where RQ allocation was 4x16 which is equal to 1 WQE BB,
causing 1 BB offset in the page and splitting the GTA WQE between different
physical pages.
The solution was to create the RQ with a even number of BBs and to have the
RQ aligned to a page.
Signed-off-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Instead of using the write buffer for reading we will use a dedicated
buffer only for reading ICM memory.
Due to the new support for args, we can have a case with pending_wc
being odd number, and with reading into the same write buffer, it is
possible to overwrite next write on the same slot.
For example:
pending_wc is 17 so the buffer for write is:
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
and we have requests as follows:
r wr wr wr wr wr wr wr wr
Now, the first read will be written into the last write because we use
the same buffer for read and write, before it was written to the HW and
we will have a wrong data in the ICM area.
Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
The accelerated modify header arguments are written in the HW area
with special WQE and specific data format.
New function was added to support writing of new argument type.
Note that GTA WQE is larger than READ and WRITE, so the queue
management logic was updated to support this.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Add functions for creation/destruction of the new type of general object.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
This way we are able to allocate chunk for modify_headers from 2 types:
STEv0 that is allocated from the action area, and STEv1 that is allocating
the chunks from the special area for patterns.
Signed-off-by: Muhammad Sammar <muhammads@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Starting with ConnectX-6 Dx, we use new design of modify_header FW object.
The current modify_header object allows for having only limited number
of FW objects, so the new design of pattern and argument allows pattern
reuse, saving memory, and having a large number of modify_header objects.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Move ACTION_CACHE_LINE_SIZE macro to header to be used by
the pattern functions as well.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
During probe, get the hardware-allowed max MTU by querying the device
configuration. Users can select MTU up to the device limit.
When XDP is in use, limit MTU settings so the buffer size is within
one page. And, when MTU is set to a too large value, XDP is not allowed
to run.
Also, to prevent changing MTU fails, and leaves the NIC in a bad state,
pre-allocate all buffers before starting the change. So in low memory
condition, it will return error, without affecting the NIC.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Update RX data path to allocate and use RX queue DMA buffers with
proper size based on potentially various MTU sizes.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Move out common buffer allocation code from mana_process_rx_cqe() and
mana_alloc_rx_wqe() to helper functions.
Refactor related variables so they can be changed in one place, and buffer
sizes are in sync.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use napi_build_skb() instead of build_skb() to take advantage of the
NAPI percpu caches to obtain skbuff_head.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5-updates-2023-04-11
1) Vlad adds the support for linux bridge multicast offload support
Patches #1 through #9
Synopsis
Vlad Says:
==============
Implement support of bridge multicast offload in mlx5. Handle port object
attribute SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED notification to toggle multicast
offload and bridge snooping support on bridge. Handle port object
SWITCHDEV_OBJ_ID_PORT_MDB notification to attach a bridge port to MDB.
Steering architecture
Existing offload infrastructure relies on two levels of flow tables - bridge
ingress and egress. For multicast offload the architecture is extended with
additional layer of per-port multicast replication tables. Such tables filter
loopback traffic (so packets are not replicated to their source port) and pop
VLAN headers for "untagged" VLANs. The tables are referenced by the MDB rules in
egress table. MDB egress rule can point to multiple per-port multicast tables,
which causes matching multicast traffic to be replicated to all of them, and,
consecutively, to several bridge ports:
+--------+--+
+---------------------------------------> Port 1 | |
| +-^------+--+
| |
| |
+-----------------------------------------+ | +---------------------------+ |
| EGRESS table | | +--> PORT 1 multicast table | |
+----------------------------------+ +-----------------------------------------+ | | +---------------------------+ |
| INGRESS table | | | | | | | |
+----------------------------------+ | dst_mac=P1,vlan=X -> pop vlan, goto P1 +--+ | | FG0: | |
| | | dst_mac=P1,vlan=Y -> pop vlan, goto P1 | | | src_port=dst_port -> drop | |
| src_mac=M1,vlan=X -> goto egress +---> dst_mac=P2,vlan=X -> pop vlan, goto P2 +--+ | | FG1: | |
| ... | | dst_mac=P2,vlan=Y -> goto P2 | | | | VLAN X -> pop, goto port | |
| | | dst_mac=MDB1,vlan=Y -> goto mcast P1,P2 +-----+ | ... | |
+----------------------------------+ | | | | | VLAN Y -> pop, goto port +-------+
+-----------------------------------------+ | | | FG3: |
| | | matchall -> goto port |
| | | |
| | +---------------------------+
| |
| |
| | +--------+--+
+---------------------------------------> Port 2 | |
| +-^------+--+
| |
| |
| +---------------------------+ |
+--> PORT 2 multicast table | |
+---------------------------+ |
| | |
| FG0: | |
| src_port=dst_port -> drop | |
| FG1: | |
| VLAN X -> pop, goto port | |
| ... | |
| | |
| FG3: | |
| matchall -> goto port +-------+
| |
+---------------------------+
Patches overview:
- Patch 1 adds hardware definition bits for capabilities required to replicate
multicast packets to multiple per-port tables. These bits are used by
following patches to only attempt multicast offload if firmware and hardware
provide necessary support.
- Pathces 2-4 patches are preparations and refactoring.
- Patch 5 implements necessary infrastructure to toggle multicast offload
via SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED port object attribute notification.
This also enabled IGMP and MLD snooping.
- Patch 6 implements per-port multicast replication tables. It only supports
filtering of loopback packets.
- Patch 7 extends per-port multicast tables with VLAN pop support for 'untagged'
VLANs.
- Patch 8 handles SWITCHDEV_OBJ_ID_PORT_MDB port object notifications. It
creates MDB replication rules in egress table that can replicate packets to
multiple per-port multicast tables.
- Patch 9 adds tracepoints for MDB events.
==============
2) Parav Create a new allocation profile for SFs, to save on memory
3) Yevgeny provides some initial patches for upcoming software steering
support new pattern/arguments type of modify_header actions.
Starting with ConnectX-6 DX, we use a new design of modify_header FW object.
The current modify_header object allows for having only limited number of
these FW objects, which means that we are limited in the number of offloaded
flows that require modify_header action.
As a preparation Yevgeny provides the following 4 patches:
- Patch 1: Add required mlx5_ifc HW bits
- Patch 2, 3: Add new WQE type and opcode that is required for pattern/arg
support and adds appropriate support in dr_send.c
- Patch 4: Add ICM pool for modify-header-pattern objects and implement
patterns cache, allowing patterns reuse for different flows
* tag 'mlx5-updates-2023-04-11' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
net/mlx5: DR, Add modify-header-pattern ICM pool
net/mlx5: DR, Prepare sending new WQE type
net/mlx5: Add new WQE for updating flow table
net/mlx5: Add mlx5_ifc bits for modify header argument
net/mlx5: DR, Set counter ID on the last STE for STEv1 TX
net/mlx5: Create a new profile for SFs
net/mlx5: Bridge, add tracepoints for multicast
net/mlx5: Bridge, implement mdb offload
net/mlx5: Bridge, support multicast VLAN pop
net/mlx5: Bridge, add per-port multicast replication tables
net/mlx5: Bridge, snoop igmp/mld packets
net/mlx5: Bridge, extract code to lookup parent bridge of port
net/mlx5: Bridge, move additional data structures to priv header
net/mlx5: Bridge, increase bridge tables sizes
net/mlx5: Add mlx5_ifc definitions for bridge multicast support
====================
Link: https://lore.kernel.org/r/20230412040752.14220-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
PFs which support the MAC Merge layer also have a set of 8 registers
called "Port traffic class N frame preemption register (PTC0FPR - PTC7FPR)".
Through these, a traffic class (group of TX rings of same dequeue
priority) can be mapped to the eMAC or to the pMAC.
There's nothing particularly spectacular here. We should probably only
commit the preemptible TCs to hardware once the MAC Merge layer became
active, but unlike Felix, we don't have an IRQ that notifies us of that.
We'd have to sleep for up to verifyTime (127 ms) to wait for a
resolution coming from the verification state machine; not only from the
ndo_setup_tc() code path, but also from enetc_mm_link_state_update().
Since it's relatively complicated and has a relatively small benefit,
I'm not doing it.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
To gain access to the larger encapsulating structure which has the type
tc_mqprio_qopt_offload, rename just the "qopt" field as "qopt".
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The seconds input from BD (6 bits) just needs to be ORed with the
upper bits from timer in this function. Avoid addition operation
every single time. Seconds rollover handling is left untouched.
Signed-off-by: Harini Katakam <harini.katakam@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Enable transmission and reception of PTP unicast packets by
updating PTP unicast config bit and setting current HW mac
address as allowed address in PTP unicast filter registers.
Signed-off-by: Harini Katakam <harini.katakam@xilinx.com>
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
There are currently two checks for PTP functionality - one on GEM
capability and another on the kernel config option. Combine them
into a single function as there's no use case where gem_has_ptp is
TRUE and MACB_USE_HWSTAMP is false.
Signed-off-by: Harini Katakam <harini.katakam@amd.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Since it is hopefully now clear that, since "last" and "layout[i].reg"
are enum types and not addresses, the existing WARN_ON() is ineffective
in checking that the _addresses_ are sorted in the proper order.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The "int i" used to index the struct ocelot_stat_layout array actually
has a specific type: enum ocelot_stat. Use it, so that the WARN()
comment from ocelot_prepare_stats_regions() makes more sense.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Use the specific enum ocelot_reg to make it clear that the region
registers are encoded and not plain addresses.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Commit a3bb8f521fd8 ("net: mscc: ocelot: remove unnecessary exposure of
stats structures") made an unnecessary change which was to add a new
line at the end of ocelot_stats.c. Remove it.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
To make it easier to debug future issues with statistics counters not
getting aggregated properly into regions, like what happened in commit
6acc72a43eac ("net: mscc: ocelot: fix stats region batching"), add some
dev_dbg() prints which show the regions that were dynamically
determined.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
ocelot_io.c duplicates the decoding of an enum ocelot_reg (which holds
an enum ocelot_target in the upper bits and an index into a regmap array
in the lower bits) 4 times.
We'd like to reuse that logic once more, from ocelot.c. In order to do
that, let's consolidate the existing 4 instances into a header
accessible both by ocelot.c as well as by ocelot_io.c.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The "u32 reg" argument that is passed to these functions is not a plain
address, but rather a driver-specific encoding of another enum
ocelot_target target in the upper bits, and an index into the
u32 ocelot->map[target][] array in the lower bits. That encoded value
takes the type "enum ocelot_reg" and is what is passed to these I/O
functions, so let's actually use that to prevent type confusion.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Conflicts:
tools/testing/selftests/net/config
62199e3f1658 ("selftests: net: Add VXLAN MDB test")
3a0385be133e ("selftests: add the missing CONFIG_IP_SCTP in net config")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Update API for bpf_xdp_metadata_rx_hash() with arg for xdp rss hash type
via matching individual Completion Queue Entry (CQE) status bits.
Fixes: ab46182d0dcb ("net/mlx4_en: Support RX XDP metadata")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/168132893562.340624.12779118462402031248.stgit@firesoul
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Update API for bpf_xdp_metadata_rx_hash() with arg for xdp rss hash type
via mapping table.
The mlx5 hardware can also identify and RSS hash IPSEC. This indicate
hash includes SPI (Security Parameters Index) as part of IPSEC hash.
Extend xdp core enum xdp_rss_hash_type with IPSEC hash type.
Fixes: bc8d405b1ba9 ("net/mlx5e: Support RX XDP metadata")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/168132892548.340624.11185734579430124869.stgit@firesoul
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
The RSS hash type specifies what portion of packet data NIC hardware used
when calculating RSS hash value. The RSS types are focused on Internet
traffic protocols at OSI layers L3 and L4. L2 (e.g. ARP) often get hash
value zero and no RSS type. For L3 focused on IPv4 vs. IPv6, and L4
primarily TCP vs UDP, but some hardware supports SCTP.
Hardware RSS types are differently encoded for each hardware NIC. Most
hardware represent RSS hash type as a number. Determining L3 vs L4 often
requires a mapping table as there often isn't a pattern or sorting
according to ISO layer.
The patch introduce a XDP RSS hash type (enum xdp_rss_hash_type) that
contains both BITs for the L3/L4 types, and combinations to be used by
drivers for their mapping tables. The enum xdp_rss_type_bits get exposed
to BPF via BTF, and it is up to the BPF-programmer to match using these
defines.
This proposal change the kfunc API bpf_xdp_metadata_rx_hash() adding
a pointer value argument for provide the RSS hash type.
Change signature for all xmo_rx_hash calls in drivers to make it compile.
The RSS type implementations for each driver comes as separate patches.
Fixes: 3d76a4d3d4e5 ("bpf: XDP metadata RX kfuncs")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/168132892042.340624.582563003880565460.stgit@firesoul
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
For quite some time we were chasing a bug which looked like a sudden
permanent failure of networking and mmc on some of our devices.
The bug was very sensitive to any software changes and even more to
any kernel debug options.
Finally we got a setup where the problem was reproducible with
CONFIG_DMA_API_DEBUG=y and it revealed the issue with the rx dma:
[ 16.992082] ------------[ cut here ]------------
[ 16.996779] DMA-API: macb ff0b0000.ethernet: device driver tries to free DMA memory it has not allocated [device address=0x0000000875e3e244] [size=1536 bytes]
[ 17.011049] WARNING: CPU: 0 PID: 85 at kernel/dma/debug.c:1011 check_unmap+0x6a0/0x900
[ 17.018977] Modules linked in: xxxxx
[ 17.038823] CPU: 0 PID: 85 Comm: irq/55-8000f000 Not tainted 5.4.0 #28
[ 17.045345] Hardware name: xxxxx
[ 17.049528] pstate: 60000005 (nZCv daif -PAN -UAO)
[ 17.054322] pc : check_unmap+0x6a0/0x900
[ 17.058243] lr : check_unmap+0x6a0/0x900
[ 17.062163] sp : ffffffc010003c40
[ 17.065470] x29: ffffffc010003c40 x28: 000000004000c03c
[ 17.070783] x27: ffffffc010da7048 x26: ffffff8878e38800
[ 17.076095] x25: ffffff8879d22810 x24: ffffffc010003cc8
[ 17.081407] x23: 0000000000000000 x22: ffffffc010a08750
[ 17.086719] x21: ffffff8878e3c7c0 x20: ffffffc010acb000
[ 17.092032] x19: 0000000875e3e244 x18: 0000000000000010
[ 17.097343] x17: 0000000000000000 x16: 0000000000000000
[ 17.102647] x15: ffffff8879e4a988 x14: 0720072007200720
[ 17.107959] x13: 0720072007200720 x12: 0720072007200720
[ 17.113261] x11: 0720072007200720 x10: 0720072007200720
[ 17.118565] x9 : 0720072007200720 x8 : 000000000000022d
[ 17.123869] x7 : 0000000000000015 x6 : 0000000000000098
[ 17.129173] x5 : 0000000000000000 x4 : 0000000000000000
[ 17.134475] x3 : 00000000ffffffff x2 : ffffffc010a1d370
[ 17.139778] x1 : b420c9d75d27bb00 x0 : 0000000000000000
[ 17.145082] Call trace:
[ 17.147524] check_unmap+0x6a0/0x900
[ 17.151091] debug_dma_unmap_page+0x88/0x90
[ 17.155266] gem_rx+0x114/0x2f0
[ 17.158396] macb_poll+0x58/0x100
[ 17.161705] net_rx_action+0x118/0x400
[ 17.165445] __do_softirq+0x138/0x36c
[ 17.169100] irq_exit+0x98/0xc0
[ 17.172234] __handle_domain_irq+0x64/0xc0
[ 17.176320] gic_handle_irq+0x5c/0xc0
[ 17.179974] el1_irq+0xb8/0x140
[ 17.183109] xiic_process+0x5c/0xe30
[ 17.186677] irq_thread_fn+0x28/0x90
[ 17.190244] irq_thread+0x208/0x2a0
[ 17.193724] kthread+0x130/0x140
[ 17.196945] ret_from_fork+0x10/0x20
[ 17.200510] ---[ end trace 7240980785f81d6f ]---
[ 237.021490] ------------[ cut here ]------------
[ 237.026129] DMA-API: exceeded 7 overlapping mappings of cacheline 0x0000000021d79e7b
[ 237.033886] WARNING: CPU: 0 PID: 0 at kernel/dma/debug.c:499 add_dma_entry+0x214/0x240
[ 237.041802] Modules linked in: xxxxx
[ 237.061637] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G W 5.4.0 #28
[ 237.068941] Hardware name: xxxxx
[ 237.073116] pstate: 80000085 (Nzcv daIf -PAN -UAO)
[ 237.077900] pc : add_dma_entry+0x214/0x240
[ 237.081986] lr : add_dma_entry+0x214/0x240
[ 237.086072] sp : ffffffc010003c30
[ 237.089379] x29: ffffffc010003c30 x28: ffffff8878a0be00
[ 237.094683] x27: 0000000000000180 x26: ffffff8878e387c0
[ 237.099987] x25: 0000000000000002 x24: 0000000000000000
[ 237.105290] x23: 000000000000003b x22: ffffffc010a0fa00
[ 237.110594] x21: 0000000021d79e7b x20: ffffffc010abe600
[ 237.115897] x19: 00000000ffffffef x18: 0000000000000010
[ 237.121201] x17: 0000000000000000 x16: 0000000000000000
[ 237.126504] x15: ffffffc010a0fdc8 x14: 0720072007200720
[ 237.131807] x13: 0720072007200720 x12: 0720072007200720
[ 237.137111] x11: 0720072007200720 x10: 0720072007200720
[ 237.142415] x9 : 0720072007200720 x8 : 0000000000000259
[ 237.147718] x7 : 0000000000000001 x6 : 0000000000000000
[ 237.153022] x5 : ffffffc010003a20 x4 : 0000000000000001
[ 237.158325] x3 : 0000000000000006 x2 : 0000000000000007
[ 237.163628] x1 : 8ac721b3a7dc1c00 x0 : 0000000000000000
[ 237.168932] Call trace:
[ 237.171373] add_dma_entry+0x214/0x240
[ 237.175115] debug_dma_map_page+0xf8/0x120
[ 237.179203] gem_rx_refill+0x190/0x280
[ 237.182942] gem_rx+0x224/0x2f0
[ 237.186075] macb_poll+0x58/0x100
[ 237.189384] net_rx_action+0x118/0x400
[ 237.193125] __do_softirq+0x138/0x36c
[ 237.196780] irq_exit+0x98/0xc0
[ 237.199914] __handle_domain_irq+0x64/0xc0
[ 237.204000] gic_handle_irq+0x5c/0xc0
[ 237.207654] el1_irq+0xb8/0x140
[ 237.210789] arch_cpu_idle+0x40/0x200
[ 237.214444] default_idle_call+0x18/0x30
[ 237.218359] do_idle+0x200/0x280
[ 237.221578] cpu_startup_entry+0x20/0x30
[ 237.225493] rest_init+0xe4/0xf0
[ 237.228713] arch_call_rest_init+0xc/0x14
[ 237.232714] start_kernel+0x47c/0x4a8
[ 237.236367] ---[ end trace 7240980785f81d70 ]---
Lars was fast to find an explanation: according to the datasheet
bit 2 of the rx buffer descriptor entry has a different meaning in the
extended mode:
Address [2] of beginning of buffer, or
in extended buffer descriptor mode (DMA configuration register [28] = 1),
indicates a valid timestamp in the buffer descriptor entry.
The macb driver didn't mask this bit while getting an address and it
eventually caused a memory corruption and a dma failure.
The problem is resolved by explicitly clearing the problematic bit
if hw timestamping is used.
Fixes: 7b4296148066 ("net: macb: Add support for PTP timestamps in DMA descriptors")
Signed-off-by: Roman Gushchin <roman.gushchin@linux.dev>
Co-developed-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20230412232144.770336-1-roman.gushchin@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Eric points out that we should make sure that ring index updates
are wrapped in the appropriate READ_ONCE/WRITE_ONCE macros.
Suggested-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Eric points out that we should make sure that ring index updates
are wrapped in the appropriate READ_ONCE/WRITE_ONCE macros.
Suggested-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Add the new programming sequence needed for EMAC3 based platforms such
as the sc8280xp family.
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
It seems that this variable should be used for all speeds, not just
1000/100.
While at it refactor it slightly to be more readable, including fixing
the typo in the variable name.
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
The driver currently sets a MAC TX delay of 2 ns no matter what the
phy-mode is. If the phy-mode indicates the phy is in charge of the
TX delay (rgmii-txid, rgmii-id), don't do it in the MAC.
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Some platforms have dwmac4 implementations that have a different
address space layout than the default, resulting in the need to define
their own DMA/MTL offsets.
Extend the functions to allow a platform driver to indicate what its
addresses are, overriding the defaults.
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Passing stmmac_priv to some of the callbacks allows hwif implementations
to grab some data that platforms can customize. Adjust the callbacks
accordingly in preparation of such a platform customization.
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
There's a few spots in the hardware interface where a void pointer is
used, but what's passed in and later cast out is always the same type.
Just use the proper type directly.
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
DAM is supposed to be DMA. Fix it to improve readability.
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
The brackets are unnecessary, remove them to match the coding style
used in the kernel.
Signed-off-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Brian Masney <bmasney@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
I have observed an issue where the RX direction of the LS1028A ENETC pMAC
seems unresponsive. The minimal procedure to reproduce the issue is:
1. Connect ENETC port 0 with a loopback RJ45 cable to one of the Felix
switch ports (0).
2. Bring the ports up (MAC Merge layer is not enabled on either end).
3. Send a large quantity of unidirectional (express) traffic from Felix
to ENETC. I tried altering frame size and frame count, and it doesn't
appear to be specific to either of them, but rather, to the quantity
of octets received. Lowering the frame count, the minimum quantity of
packets to reproduce relatively consistently seems to be around 37000
frames at 1514 octets (w/o FCS) each.
4. Using ethtool --set-mm, enable the pMAC in the Felix and in the ENETC
ports, in both RX and TX directions, and with verification on both
ends.
5. Wait for verification to complete on both sides.
6. Configure a traffic class as preemptible on both ends.
7. Send some packets again.
The issue is at step 5, where the verification process of ENETC ends
(meaning that Felix responds with an SMD-R and ENETC sees the response),
but the verification process of Felix never ends (it remains VERIFYING).
If step 3 is skipped or if ENETC receives less traffic than
approximately that threshold, the test runs all the way through
(verification succeeds on both ends, preemptible traffic passes fine).
If, between step 4 and 5, the step below is also introduced:
4.1. Disable and re-enable PM0_COMMAND_CONFIG bit RX_EN
then again, the sequence of steps runs all the way through, and
verification succeeds, even if there was the previous RX traffic
injected into ENETC.
Traffic sent *by* the ENETC port prior to enabling the MAC Merge layer
does not seem to influence the verification result, only received
traffic does.
The LS1028A manual does not mention any relationship between
PM0_COMMAND_CONFIG and MMCSR, and the hardware people don't seem to
know for now either.
The bit that is toggled to work around the issue is also toggled
by enetc_mac_enable(), called from phylink's mac_link_down() and
mac_link_up() methods - which is how the workaround was found:
verification would work after a link down/up.
Fixes: c7b9e8086902 ("net: enetc: add support for MAC Merge layer")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20230411192645.1896048-1-vladimir.oltean@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Remove an inability of bnxt_en driver to set eswitch to switchdev
mode without existing VFs by:
1. Allow to set switchdev mode in bnxt_dl_eswitch_mode_set() so
representors are created only when num_vfs > 0 otherwise just
set bp->eswitch_mode
2. Do not automatically change bp->eswitch_mode during
bnxt_vf_reps_create() and bnxt_vf_reps_destroy() calls so
the eswitch mode is managed only by an user by devlink.
Just set temporarily bp->eswitch_mode to legacy to avoid
re-opening of representors during destroy.
3. Create representors in bnxt_sriov_enable() if current eswitch
mode is switchdev one
Tested by this sequence:
1. Set PF interface up
2. Set PF's eswitch mode to switchdev
3. Created N VFs
4. Checked that N representors were created
5. Set eswitch mode to legacy
6. Checked that representors were deleted
7. Set eswitch mode back to switchdev
8. Checked that representors exist again for VFs
9. Deleted all VFs
10. Checked that all representors were deleted as well
11. Checked that current eswitch mode is still switchdev
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Venkat Duvvuru <venkatkumar.duvvuru@broadcom.com>
Link: https://lore.kernel.org/r/20230411120443.126055-1-ivecera@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|