| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
There is a race between reset and the transmit paths that can lead to
ibmvnic_xmit() accessing an scrq after it has been freed in the reset
path. It can result in a crash like:
Kernel attempted to read user page (0) - exploit attempt? (uid: 0)
BUG: Kernel NULL pointer dereference on read at 0x00000000
Faulting instruction address: 0xc0080000016189f8
Oops: Kernel access of bad area, sig: 11 [#1]
...
NIP [c0080000016189f8] ibmvnic_xmit+0x60/0xb60 [ibmvnic]
LR [c000000000c0046c] dev_hard_start_xmit+0x11c/0x280
Call Trace:
[c008000001618f08] ibmvnic_xmit+0x570/0xb60 [ibmvnic] (unreliable)
[c000000000c0046c] dev_hard_start_xmit+0x11c/0x280
[c000000000c9cfcc] sch_direct_xmit+0xec/0x330
[c000000000bfe640] __dev_xmit_skb+0x3a0/0x9d0
[c000000000c00ad4] __dev_queue_xmit+0x394/0x730
[c008000002db813c] __bond_start_xmit+0x254/0x450 [bonding]
[c008000002db8378] bond_start_xmit+0x40/0xc0 [bonding]
[c000000000c0046c] dev_hard_start_xmit+0x11c/0x280
[c000000000c00ca4] __dev_queue_xmit+0x564/0x730
[c000000000cf97e0] neigh_hh_output+0xd0/0x180
[c000000000cfa69c] ip_finish_output2+0x31c/0x5c0
[c000000000cfd244] __ip_queue_xmit+0x194/0x4f0
[c000000000d2a3c4] __tcp_transmit_skb+0x434/0x9b0
[c000000000d2d1e0] __tcp_retransmit_skb+0x1d0/0x6a0
[c000000000d2d984] tcp_retransmit_skb+0x34/0x130
[c000000000d310e8] tcp_retransmit_timer+0x388/0x6d0
[c000000000d315ec] tcp_write_timer_handler+0x1bc/0x330
[c000000000d317bc] tcp_write_timer+0x5c/0x200
[c000000000243270] call_timer_fn+0x50/0x1c0
[c000000000243704] __run_timers.part.0+0x324/0x460
[c000000000243894] run_timer_softirq+0x54/0xa0
[c000000000ea713c] __do_softirq+0x15c/0x3e0
[c000000000166258] __irq_exit_rcu+0x158/0x190
[c000000000166420] irq_exit+0x20/0x40
[c00000000002853c] timer_interrupt+0x14c/0x2b0
[c000000000009a00] decrementer_common_virt+0x210/0x220
--- interrupt: 900 at plpar_hcall_norets_notrace+0x18/0x2c
The immediate cause of the crash is the access of tx_scrq in the following
snippet during a reset, where the tx_scrq can be either NULL or an address
that will soon be invalid:
ibmvnic_xmit()
{
...
tx_scrq = adapter->tx_scrq[queue_num];
txq = netdev_get_tx_queue(netdev, queue_num);
ind_bufp = &tx_scrq->ind_buf;
if (test_bit(0, &adapter->resetting)) {
...
}
But beyond that, the call to ibmvnic_xmit() itself is not safe during a
reset and the reset path attempts to avoid this by stopping the queue in
ibmvnic_cleanup(). However just after the queue was stopped, an in-flight
ibmvnic_complete_tx() could have restarted the queue even as the reset is
progressing.
Since the queue was restarted we could get a call to ibmvnic_xmit() which
can then access the bad tx_scrq (or other fields).
We cannot however simply have ibmvnic_complete_tx() check the ->resetting
bit and skip starting the queue. This can race at the "back-end" of a good
reset which just restarted the queue but has not cleared the ->resetting
bit yet. If we skip restarting the queue due to ->resetting being true,
the queue would remain stopped indefinitely potentially leading to transmit
timeouts.
IOW ->resetting is too broad for this purpose. Instead use a new flag
that indicates whether or not the queues are active. Only the open/
reset paths control when the queues are active. ibmvnic_complete_tx()
and others wake up the queue only if the queue is marked active.
So we will have:
A. reset/open thread in ibmvnic_cleanup() and __ibmvnic_open()
->resetting = true
->tx_queues_active = false
disable tx queues
...
->tx_queues_active = true
start tx queues
B. Tx interrupt in ibmvnic_complete_tx():
if (->tx_queues_active)
netif_wake_subqueue();
To ensure that ->tx_queues_active and state of the queues are consistent,
we need a lock which:
- must also be taken in the interrupt path (ibmvnic_complete_tx())
- shared across the multiple queues in the adapter (so they don't
become serialized)
Use rcu_read_lock() and have the reset thread synchronize_rcu() after
updating the ->tx_queues_active state.
While here, consolidate a few boolean fields in ibmvnic_adapter for
better alignment.
Based on discussions with Brian King and Dany Madden.
Fixes: 7ed5b31f4a66 ("net/ibmvnic: prevent more than one thread from running in reset")
Reported-by: Vaishnavi Bhat <vaish123@in.ibm.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Add support for PTP-IO Event Output (Periodic Output - perout) for
PCI11010/PCI11414 chips
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
PTP-IOs block provides for time stamping PTP-IO input events.
PTP-IOs are numbered from 0 to 11.
When a PTP-IO is enabled by the corresponding bit in the PTP-IO
Capture Configuration Register, a rising or falling edge,
respectively, will capture the 1588 Local Time Counter
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com>
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Add new the OTP read and write access functions for PCI11010/PCI11414 chips
PCI11010/PCI11414 OTP module register offsets are different from
LAN743x OTP module
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Add new the EEPROM read and write access functions and system lock
protection to access by devices for PCI11010/PCI11414 chips
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Tx 4 queue statistics display through ethtool application
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Saeed Mahameed says:
====================
mlx5-updates-2022-03-17
1) From Maxim Mikityanskiy,
Datapath improvements in preparation for XDP multi buffer
This series contains general improvements for the datapath that are
useful for the upcoming XDP multi buffer support:
a. Non-linear legacy RQ: validate MTU for robustness, build the linear
part of SKB over the first hardware fragment (instead of copying the
packet headers), adjust headroom calculations to allow enabling headroom
in the non-linear mode (useful for XDP multi buffer).
b. XDP: do the XDP program test before function call, optimize
parameters of mlx5e_xdp_handle.
2) From Rongwei Liu, DR, reduce steering memory usage
Currently, mlx5 driver uses mlx5_htbl/chunk/ste to organize
steering logic. However there is a little memory waste.
This update targets to reduce steering memory footprint by:
a. Adjust struct member layout.
b. Remove duplicated indicator by using simple functions call.
With 500k TX rules(3 ste) plus 500k RX rules(6 stes), these patches
can save around 17% memory.
3) Three cleanup commits at the end of this series.
===================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Gain support for port mirroring using tc-matchall by forwarding the
calls to the ocelot switch library.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Drivers might have error messages to propagate to user space, most
common being that they support a single mirror port.
Propagate the netlink extack so that they can inform user space in a
verbal way of their limitations.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Per-flow mirroring with the VCAP IS2 TCAM (in itself handled as an
offload for tc-flower) is done by setting the MIRROR_ENA bit from the
action vector of the filter. The packet is mirrored to the port mask
configured in the ANA:ANA:MIRRORPORTS register (the same port mask as
the destinations for port-based mirroring).
Functionality was tested with:
tc qdisc add dev swp3 clsact
tc filter add dev swp3 ingress protocol ip \
flower skip_sw ip_proto icmp \
action mirred egress mirror dev swp1
and pinging through swp3, while seeing that the ICMP replies are
mirrored towards swp1.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Some VCAP filters utilize resources which are global to the switch, like
for example VCAP IS2 policers take an index into a global policer pool.
In commit c9a7fe1238e5 ("net: mscc: ocelot: add action of police on
vcap_is2"), Xiaoliang expressed this by hooking into the low-level
ocelot_vcap_filter_add_to_block() and ocelot_vcap_block_remove_filter()
functions, and allocating/freeing the policers from there.
Evaluating the code, there probably isn't a better place, but we'll need
to do something similar for the mirror ports, and the code will start to
look even more hacked up than it is right now.
Create two ocelot_vcap_filter_{add,del}_aux_resources() functions to
contain the madness, and pollute less the body of other functions such
as ocelot_vcap_filter_add_to_block().
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Ocelot switches perform port-based ingress mirroring if
ANA:PORT:PORT_CFG field SRC_MIRROR_ENA is set, and egress mirroring if
the port is in ANA:ANA:EMIRRORPORTS.
Both ingress-mirrored and egress-mirrored frames are copied to the port
mask from ANA:ANA:MIRRORPORTS.
So the choice of limiting to a single mirror port via ocelot_mirror_get()
and ocelot_mirror_put() may seem bizarre, but the hardware model doesn't
map very well to the user space model. If the user wants to mirror the
ingress of swp1 towards swp2 and the ingress of swp3 towards swp4, we'd
have to program ANA:ANA:MIRRORPORTS with BIT(2) | BIT(4), and that would
make swp1 be mirrored towards swp4 too, and swp3 towards swp2. But there
are no tc-matchall rules to describe those actions.
Now, we could offload a matchall rule with multiple mirred actions, one
per desired mirror port, and force the user to stick to the multi-action
rule format for subsequent matchall filters. But both DSA and ocelot
have the flow_offload_has_one_action() check for the matchall offload,
plus the fact that it will get cumbersome to cross-check matchall
mirrors with flower mirrors (which will be added in the next patch).
As a result, we limit the configuration to a single mirror port, with
the possibility of lifting the restriction in the future.
Frames injected from the CPU don't get egress-mirrored, since they are
sent with the BYPASS bit in the injection frame header, and this
bypasses the analyzer module (effectively also the mirroring logic).
I don't know what to do/say about this.
Functionality was tested with:
tc qdisc add dev swp3 clsact
tc filter add dev swp3 ingress \
matchall skip_sw \
action mirred egress mirror dev swp1
and pinging through swp3, while seeing that the ICMP replies are
mirrored towards swp1.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In preparation for adding port mirroring support to the ocelot driver,
the dispatching function ocelot_setup_tc_cls_matchall() must be free of
action-specific code. Move port policer creation and deletion to
separate functions.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
According to erratum described in DS80000687C[1]: "Module 2: Link drops with
some EEE link partners.", we need to "Disable the EEE next page
exchange in EEE Global Register 2"
1 - https://ww1.microchip.com/downloads/en/DeviceDoc/KSZ87xx-Errata-DS80000687C.pdf
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20220316125529.1489045-1-o.rempel@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Allocate a SID in the STU for each MSTID in use by a bridge and handle
the mapping of MSTIDs to VLANs using the SID field of each VTU entry.
Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Export the raw STU data in a devlink region so that it can be
inspected from userspace and compared to the current bridge
configuration.
Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In early LinkStreet silicon (e.g. 6095/6185), the per-VLAN STP states
were kept in the VTU - there was no concept of a SID. Later, the
information was split into two tables, where the VTU only tracked
memberships and deferred the STP state tracking to the STU via a
pointer (SID). This meant that a group of VLANs could share the same
STU entry. Most likely, this was done to align with MSTP (802.1Q-2018,
Clause 13), which is built on this principle.
While the VTU is still 4k lines on most devices, the STU is capped at
64 entries. This means that the current stategy, updating STU info
whenever a VTU entry is updated, can not easily support MSTP because:
- The maximum number of VIDs would also be capped at 64, as we would
have to allocate one SID for every VTU entry - even if many VLANs
would effectively share the same MST.
- MSTP updates would be unnecessarily slow as you would have to
iterate over all VLANs that share the same MST.
In order to support MSTP offloading in the future, manage the STU as a
separate entity from the VTU.
Only add support for newer hardware with separate VTU and
STU. VTU-only devices can also be supported, but essentially this
requires a software implementation of an STU (fanning out state
changed to all VLANs tied to the same MST).
Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
There's a number of systems supporting DASH remote management.
Driver unload and system shutdown can result in the PHY suspending,
thus making DASH unusable. Improve this by handling DASH being enabled
very similar to WoL being enabled.
Tested-by: Yanko Kaneti <yaneti@declera.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://lore.kernel.org/r/1de3b176-c09c-1654-6f00-9785f7a4f954@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Tony Nguyen says:
====================
100GbE Intel Wired LAN Driver Updates 2022-03-16
This series contains updates to gtp and ice driver.
Wojciech fixes smatch reported inconsistent indenting for gtp and ice.
Yang Yingliang fixes a couple of return value checks for GNSS to IS_PTR
instead of null.
Jacob adds support for trace events on tx timestamps.
* '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue:
ice: add trace events for tx timestamps
ice: fix return value check in ice_gnss.c
ice: Fix inconsistent indenting in ice_switch
gtp: Fix inconsistent indenting
====================
Link: https://lore.kernel.org/r/20220316204024.3201500-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
There is a spelling mistake in a dev_err message. Fix it.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20220316234620.55885-1-colin.i.king@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
There is a spelling mistake in a dev_err message and the MAX_SKB_FRAGS
value does not need to be printed between parentheses. Fix this.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20220316233455.54541-1-colin.i.king@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When compiling with -Wformat, clang emits the following warning:
drivers/net/ethernet/freescale/xgmac_mdio.c:243:22: warning: format
specifies type 'unsigned char' but the argument has type 'int'
[-Wformat]
phy_id, dev_addr, regnum);
^~~~~~
./include/linux/dev_printk.h:163:47: note: expanded from macro 'dev_dbg'
dev_printk(KERN_DEBUG, dev, dev_fmt(fmt), ##__VA_ARGS__); \
~~~ ^~~~~~~~~~~
./include/linux/dev_printk.h:129:34: note: expanded from macro 'dev_printk'
_dev_printk(level, dev, fmt, ##__VA_ARGS__); \
~~~ ^~~~~~~~~~~
The types of these arguments are unconditionally defined, so this patch
updates the format character to the correct ones for ints and unsigned
ints.
Link: https://github.com/ClangBuiltLinux/linux/issues/378
Signed-off-by: Bill Wendling <morbo@google.com>
Link: https://lore.kernel.org/r/20220316213114.2352352-1-morbo@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When compiling with -Wformat, clang emits the following warnings:
drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c:6181:40: warning: format
specifies type 'unsigned short' but the argument has type 'u32'
(aka 'unsigned int') [-Wformat]
ret = scnprintf(str, *len, "%hx.%hx", num >> 16, num);
~~~ ^~~~~~~~~
%x
drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c:6181:51: warning: format
specifies type 'unsigned short' but the argument has type 'u32'
(aka 'unsigned int') [-Wformat]
ret = scnprintf(str, *len, "%hx.%hx", num >> 16, num);
~~~ ^~~
%x
drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c:6196:47: warning: format
specifies type 'unsigned char' but the argument has type 'u32'
(aka 'unsigned int') [-Wformat]
ret = scnprintf(str, *len, "%hhx.%hhx.%hhx", num >> 16, num >> 8, num);
~~~~ ^~~~~~~~~
%x
drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c:6196:58: warning: format
specifies type 'unsigned char' but the argument has type 'u32'
(aka 'unsigned int') [-Wformat]
ret = scnprintf(str, *len, "%hhx.%hhx.%hhx", num >> 16, num >> 8, num);
~~~~ ^~~~~~~~
%x
drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c:6196:68: warning: format
specifies type 'unsigned char' but the argument has type 'u32'
(aka 'unsigned int') [-Wformat]
ret = scnprintf(str, *len, "%hhx.%hhx.%hhx", num >> 16, num >> 8, num);
~~~~ ^~~
%x
The types of these arguments are unconditionally defined, so this patch
updates the format character to the correct ones for ints and unsigned
ints.
Link: https://github.com/ClangBuiltLinux/linux/issues/378
Signed-off-by: Bill Wendling <morbo@google.com>
Link: https://lore.kernel.org/r/20220316213104.2351651-1-morbo@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When compiling with -Wformat, clang emits the following warning:
drivers/net/ethernet/freescale/enetc/enetc_mdio.c:151:22: warning:
format specifies type 'unsigned char' but the argument has type 'int'
[-Wformat]
phy_id, dev_addr, regnum);
^~~~~~
./include/linux/dev_printk.h:163:47: note: expanded from macro 'dev_dbg'
dev_printk(KERN_DEBUG, dev, dev_fmt(fmt), ##__VA_ARGS__); \
~~~ ^~~~~~~~~~~
./include/linux/dev_printk.h:129:34: note: expanded from macro 'dev_printk'
_dev_printk(level, dev, fmt, ##__VA_ARGS__); \
~~~ ^~~~~~~~~~~
The types of these arguments are unconditionally defined, so this patch
updates the format character to the correct ones for ints and unsigned
ints.
Link: https://github.com/ClangBuiltLinux/linux/issues/378
Signed-off-by: Bill Wendling <morbo@google.com>
Reviewed-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Link: https://lore.kernel.org/r/20220316213109.2352015-1-morbo@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
No conflicts.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Allow increasing the MTU over page boundaries on veth devices
if the attached xdp program declares to support xdp fragments.
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/d5dc039c3d4123426e7023a488c449181a7bc57f.1646989407.git.lorenzo@kernel.org
|
|
Introduce veth_convert_skb_to_xdp_buff routine in order to
convert a non-linear skb into a xdp buffer. If the received skb
is cloned or shared, veth_convert_skb_to_xdp_buff will copy it
in a new skb composed by order-0 pages for the linear and the
fragmented area. Moreover veth_convert_skb_to_xdp_buff guarantees
we have enough headroom for xdp.
This is a preliminary patch to allow attaching xdp programs with frags
support on veth devices.
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/8d228b106bc1903571afd1d77e797bffe9a5ea7c.1646989407.git.lorenzo@kernel.org
|
|
Even if this is a theoretical issue since it is not possible to perform
XDP_REDIRECT on a non-linear xdp_frame, veth driver does not account
paged area in ndo_xdp_xmit function pointer.
Introduce xdp_get_frame_len utility routine to get the xdp_frame full
length and account total frame size running XDP_REDIRECT of a
non-linear xdp frame into a veth device.
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Toke Hoiland-Jorgensen <toke@redhat.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/54f9fd3bb65d190daf2c0bbae2f852ff16cfbaa0.1646989407.git.lorenzo@kernel.org
|
|
mlx5_fill_page_array API function is not used.
Remove it, reduce the number of exported functions.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
All WQ types moved to using the fragmented allocation API
for coherent memory. Contiguous API is not used anymore.
Remove it, reduce the number of exported functions.
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
On tuple offload del command, tuples are tried to be removed twice
from the hashtable, once directly via mlx5_tc_ct_entry_remove_from_tuples()
and a second time in the following mlx5_tc_ct_entry_put()->
mlx5_tc_ct_entry_del()->mlx5_tc_ct_entry_remove_from_tuples() call.
This doesn't cause any issue since rhashtable first checks if the
removed object exists in the hashtable.
Remove the extra mlx5_tc_ct_entry_remove_from_tuples().
Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
It can be calculated via function mlx5dr_ste_get_hw_ste().
Very simple and lightweight, no need to use a dedicated member.
Reduce 8 bytes from struct mlx5dr_ste and its size is 48 bytes now.
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Reviewed-by: Shun Hao <shunh@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Remove chunk_size in struct mlx5dr_icm_chunk and use
chunk->size instead.
Remove ste_arr/hw_ste_arr/miss_list since they can be accessed
from htbl->chunk pointer, no need to keep a copy.
This commit reduces 28 bytes from struct mlx5dr_ste_htbl and its
size is 32 bytes now.
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Reviewed-by: Shun Hao <shunh@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Target to reduce the memory consumption in large scale of flow rules.
They can be calculated quickly from buddy memory pool.
1. num_of_entries calls dr_icm_pool_get_chunk_num_of_entries().
2. byte_size calls dr_icm_pool_get_chunk_byte_size().
Use chunk size in dr_icm_chunk to speed up and the one in dr_ste_htbl
will be removed in the upcoming commit.
This commit reduce 8 bytes from struct mlx5_dr_icm_chunk and its
current size is 56 bytes.
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Reviewed-by: Shun Hao <shunh@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
It can be calculated quickly from buddy memory pool by
function mlx5dr_icm_pool_get_chunk_icm_addr().
This function is very lightweight and straightforward.
Reduce 8 bytes and current size of struct mlx5_dr_icm_chunk
is 64 bytes.
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Reviewed-by: Shun Hao <shunh@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Reduce memory footprint by removing mr_addr and rkey from
mlx5_dr_icm_chunk.
1. mr_addr is calculated by mlx5dr_icm_pool_get_chunk_mr_addr()
2. rkey is calculated by mlx5dr_icm_pool_get_chunk_rkey()
The two new functions are very lightweight and straightforward.
Reduce 8 bytes from struct mlx5_dr_icm_chunk, its current size is
72 bytes.
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Reviewed-by: Shun Hao <shunh@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Accord to profiling, mlx5dr_ste/mlx5dr_icm_chunk are the two
hot structures. Their memory layout can be optimized by
adjusting member sequences.
Struct mlx5dr_ste size changes from 64 bytes to 56 bytes.
In the upcoming commits, struct mlx5dr_icm_chunk memory layout
will change automatically after removing some members.
Keep it untouched here.
Signed-off-by: Rongwei Liu <rongweil@nvidia.com>
Reviewed-by: Shun Hao <shunh@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
The packet size in mlx5e_skb_from_cqe_mpwrq_linear can't overflow u16,
since the maximum packet size in linear striding RQ is 2^13 bytes. Drop
the unneeded u32 variable.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
The len parameter of mlx5e_xdp_handle is used to output the new packet
length after XDP has processed the packet and returned XDP_PASS.
However, this value can be calculated on the caller site, as the caller
knows if it was an XDP_PASS.
This commit drops the len parameter and moves the calculation to the
caller, reducing the number of parameters passed to the function and
preparing for XDP support in non-linear legacy RQ.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Instead of early return inside mlx5e_xdp_handle(), let the caller check
if an XDP program is loaded. This allows saving a few unnecessary
function calls and calculations in case !prog.
Performance test: single core, drop packets in iptables
Before: 3,872,504 pps
After: 3,975,628 pps (+2.66%)
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
As a performance optimization and preparation to enabling XDP multi
buffer on non-linear legacy RQ, build the linear part of the SKB over
the first fragment, instead of allocating a new buffer and copying the
first 256 bytes there.
To achieve this, add headroom and tailroom to the first fragment.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Currently, rq->buff.headroom is applied to all fragments in legacy RQ.
In the linear mode, there is a non-zero headroom, but there is only one
fragment per packet. In the non-linear mode, the headroom is zero.
This commit changes the logic to apply the headroom only to the first
fragment. The current behavior remains the same for both linear and
non-linear modes. However, it allows the next commit to enable headroom
for the non-linear mode, which will be applied only to the first
fragment.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
mlx5e_build_rq_frags_info() assumes that MTU is not bigger than
PAGE_SIZE * MLX5E_MAX_RX_FRAGS, which is 16K for 4K pages. Currently,
the firmware limits MTU to 10K, so the assumption doesn't lead to a bug.
This commits adds an additional driver check for reliability, since the
firmware boundary might be changed.
The calculation is taken to a separate function with a comment
explaining it. It's a preparation for the following patches that
introcuce XDP multi buffer support.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
|
|
Recent commit 974578017fc1 ("iavf: Add waiting so the port is
initialized in remove") adds a wait-loop at the beginning of
iavf_remove() to ensure that port initialization is finished
prior unregistering net device. This causes a regression
in reboot/shutdown scenario because in this case callback
iavf_shutdown() is called and this callback detaches the device,
makes it down if it is running and sets its state to __IAVF_REMOVE.
Later shutdown callback of associated PF driver (e.g. ice_shutdown)
is called. That callback calls among other things sriov_disable()
that calls indirectly iavf_remove() (see stack trace below).
As the adapter state is already __IAVF_REMOVE then the mentioned
loop is end-less and shutdown process hangs.
The patch fixes this by checking adapter's state at the beginning
of iavf_remove() and skips the rest of the function if the adapter
is already in remove state (shutdown is in progress).
Reproducer:
1. Create VF on PF driven by ice or i40e driver
2. Ensure that the VF is bound to iavf driver
3. Reboot
[52625.981294] sysrq: SysRq : Show Blocked State
[52625.988377] task:reboot state:D stack: 0 pid:17359 ppid: 1 f2
[52625.996732] Call Trace:
[52625.999187] __schedule+0x2d1/0x830
[52626.007400] schedule+0x35/0xa0
[52626.010545] schedule_hrtimeout_range_clock+0x83/0x100
[52626.020046] usleep_range+0x5b/0x80
[52626.023540] iavf_remove+0x63/0x5b0 [iavf]
[52626.027645] pci_device_remove+0x3b/0xc0
[52626.031572] device_release_driver_internal+0x103/0x1f0
[52626.036805] pci_stop_bus_device+0x72/0xa0
[52626.040904] pci_stop_and_remove_bus_device+0xe/0x20
[52626.045870] pci_iov_remove_virtfn+0xba/0x120
[52626.050232] sriov_disable+0x2f/0xe0
[52626.053813] ice_free_vfs+0x7c/0x340 [ice]
[52626.057946] ice_remove+0x220/0x240 [ice]
[52626.061967] ice_shutdown+0x16/0x50 [ice]
[52626.065987] pci_device_shutdown+0x34/0x60
[52626.070086] device_shutdown+0x165/0x1c5
[52626.074011] kernel_restart+0xe/0x30
[52626.077593] __do_sys_reboot+0x1d2/0x210
[52626.093815] do_syscall_64+0x5b/0x1a0
[52626.097483] entry_SYSCALL_64_after_hwframe+0x65/0xca
Fixes: 974578017fc1 ("iavf: Add waiting so the port is initialized in remove")
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Link: https://lore.kernel.org/r/20220317104524.2802848-1-ivecera@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
ACL rules can be offloaded to VCAP IS2 either through chain 0, or, since
the blamed commit, through a chain index whose number encodes a specific
PAG (Policy Action Group) and lookup number.
The chain number is translated through ocelot_chain_to_pag() into a PAG,
and through ocelot_chain_to_lookup() into a lookup number.
The problem with the blamed commit is that the above 2 functions don't
have special treatment for chain 0. So ocelot_chain_to_pag(0) returns
filter->pag = 224, which is in fact -32, but the "pag" field is an u8.
So we end up programming the hardware with VCAP IS2 entries having a PAG
of 224. But the way in which the PAG works is that it defines a subset
of VCAP IS2 filters which should match on a packet. The default PAG is
0, and previous VCAP IS1 rules (which we offload using 'goto') can
modify it. So basically, we are installing filters with a PAG on which
no packet will ever match. This is the hardware equivalent of adding
filters to a chain which has no 'goto' to it.
Restore the previous functionality by making ACL filters offloaded to
chain 0 go to PAG 0 and lookup number 0. The choice of PAG is clearly
correct, but the choice of lookup number isn't "as before" (which was to
leave the lookup a "don't care"). However, lookup 0 should be fine,
since even though there are ACL actions (policers) which have a
requirement to be used in a specific lookup, that lookup is 0.
Fixes: 226e9cd82a96 ("net: mscc: ocelot: only install TCAM entries into a specific lookup and PAG")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/20220316192117.2568261-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The RXCHK block will return a partial checksum of 0 if it encounters
a problem while receiving a packet. Since a 1's complement sum can
only produce this result if no bits are set in the received data
stream it is fair to treat it as an invalid partial checksum and
not pass it up the stack.
Fixes: 810155397890 ("net: bcmgenet: use CHECKSUM_COMPLETE for NETIF_F_RXCSUM")
Signed-off-by: Doug Berger <opendmb@gmail.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20220317012812.1313196-1-opendmb@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Commit b7a49f73059f ("bnx2x: Utilize firmware 7.13.21.0")
added request_firmware() logic in probe() which caused
load failure when firmware file is not present in initrd (below),
as access to firmware file is not feasible during probe.
Direct firmware load for bnx2x/bnx2x-e2-7.13.15.0.fw failed with error -2
Direct firmware load for bnx2x/bnx2x-e2-7.13.21.0.fw failed with error -2
This patch fixes this issue by -
1. Removing request_firmware() logic from the probe()
such that .ndo_open() handle it as it used to handle
it earlier
2. Given request_firmware() is removed from probe(), so
driver has to relax FW version comparisons a bit against
the already loaded FW version (by some other PFs of same
adapter) to allow different compatible/close enough FWs with which
multiple PFs may run with (in different environments), as the
given PF who is in probe flow has no idea now with which firmware
file version it is going to initialize the device in ndo_open()
Link: https://lore.kernel.org/all/46f2d9d9-ae7f-b332-ddeb-b59802be2bab@molgen.mpg.de/
Reported-by: Paul Menzel <pmenzel@molgen.mpg.de>
Tested-by: Paul Menzel <pmenzel@molgen.mpg.de>
Fixes: b7a49f73059f ("bnx2x: Utilize firmware 7.13.21.0")
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Ariel Elior <aelior@marvell.com>
Link: https://lore.kernel.org/r/20220316214613.6884-1-manishc@marvell.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The driver requires firmware so define MODULE_FIRMWARE so that modinfo
provides the details.
Fixes: fa164e40c53b ("net: phy: mscc: split the driver into separate files")
Signed-off-by: Juerg Haefliger <juergh@canonical.com>
Link: https://lore.kernel.org/r/20220316151835.88765-1-juergh@canonical.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Clang static analysis reports this representative issue
igb_ptp.c:997:3: warning: The left operand of '+' is a
garbage value
ktime_add_ns(shhwtstamps.hwtstamp, adjust);
^ ~~~~~~~~~~~~~~~~~~~~
shhwtstamps.hwtstamp is set by a call to
igb_ptp_systim_to_hwtstamp(). In the switch-statement
for the hw type, the hwtstamp is zeroed for matches
but not the default case. Move the memset out of
switch-statement. This degarbages the default case
and reduces the size.
Some whitespace cleanup of empty lines
Signed-off-by: Tom Rix <trix@redhat.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
|
The calculation of the checksum can fail.
So move converting the checksum to little endian
to inside the return status check.
Signed-off-by: Tom Rix <trix@redhat.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
|
Sukadev Bhattiprolu
Raju Lakkaraju
David S. Miller
Vladimir Oltean
Oleksij Rempel
Tobias Waldekranz
Heiner Kallweit
Jakub Kicinski
Colin Ian King
Bill Wendling
Lorenzo Bianconi
Tariq Toukan
Paul Blakey
Rongwei Liu
Maxim Mikityanskiy
Ivan Vecera
Doug Berger
Manish Chopra
Juerg Haefliger
Tom Rix