The following pull-request contains BPF updates for your *net-next* tree.
We've added 21 non-merge commits during the last 18 day(s) which contain
a total of 21 files changed, 1185 insertions(+), 127 deletions(-).
The main changes are:
1) Put xsk sockets on a struct diet and add various cleanups. Overall, this helps
to bump performance by 12% for some workloads, from Maciej Fijalkowski.
2) Extend BPF selftests to increase coverage of XDP features in combination
with BPF cpumap, from Alexis Lothoré (eBPF Foundation).
3) Extend netkit with an option to delegate skb->{mark,priority} scrubbing to
its BPF program, from Daniel Borkmann.
4) Make the bpf_get_netns_cookie() helper available also to tc(x) BPF programs,
from Mahe Tardy.
5) Extend BPF selftests covering a BPF program setting socket options per MPTCP
subflow, from Geliang Tang and Nicolas Rybowski.
bpf-next-for-netdev
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (21 commits)
xsk: Use xsk_buff_pool directly for cq functions
xsk: Wrap duplicated code to function
xsk: Carry a copy of xdp_zc_max_segs within xsk_buff_pool
xsk: Get rid of xdp_buff_xsk::orig_addr
xsk: s/free_list_node/list_node/
xsk: Get rid of xdp_buff_xsk::xskb_list_node
selftests/bpf: check program redirect in xdp_cpumap_attach
selftests/bpf: make xdp_cpumap_attach keep redirect prog attached
selftests/bpf: fix bpf_map_redirect call for cpu map test
selftests/bpf: add tcx netns cookie tests
bpf: add get_netns_cookie helper to tc programs
selftests/bpf: add missing header include for htons
selftests/bpf: Extend netkit tests to validate skb meta data
tools: Sync if_link.h uapi tooling header
netkit: Add add netkit scrub support to rt_link.yaml
netkit: Simplify netkit mode over to use NLA_POLICY_MAX
netkit: Add option for scrubbing skb meta data
bpf: Remove unused macro
selftests/bpf: Add mptcp subflow subtest
selftests/bpf: Add getsockopt to inspect mptcp subflow
...
====================
Jakub Kicinski [Thu, 10 Oct 2024 15:12:48 +0000 (08:12 -0700)]
tools: ynl-gen: use names of constants in generated limits
YNL specs can use string expressions for limits, like s32-min
or u16-max. We convert all of those into their numeric values
when generating the code, which isn't always helpful. Try to
retain the string representations in the output. Any sort of
calculations still need the integers.
Daniel Golle [Thu, 10 Oct 2024 12:55:17 +0000 (13:55 +0100)]
net: phy: mxl-gpy: correctly describe LED polarity
According the datasheet covering the LED (0x1b) register:
0B Active High LEDx pin driven high when activated
1B Active Low LEDx pin driven low when activated
Make use of the now available 'active-high' property and correctly
reflect the polarity setting which was previously inverted.
Daniel Golle [Thu, 10 Oct 2024 12:55:00 +0000 (13:55 +0100)]
net: phy: aquantia: correctly describe LED polarity override
Use newly defined 'active-high' property to set the
VEND1_GLOBAL_LED_DRIVE_VDD bit and let 'active-low' clear that bit. This
reflects the technical reality which was inverted in the previous
description in which the 'active-low' property was used to actually set
the VEND1_GLOBAL_LED_DRIVE_VDD bit, which means that VDD (ie. supply
voltage) of the LED is driven rather than GND.
Daniel Golle [Thu, 10 Oct 2024 12:54:19 +0000 (13:54 +0100)]
net: phy: support 'active-high' property for PHY LEDs
In addition to 'active-low' and 'inactive-high-impedance' also
support 'active-high' property for PHY LED pin configuration.
As only either 'active-high' or 'active-low' can be set at the
same time, WARN and return an error in case both are set.
====================
make PHY output RMII reference clock
The TJA11xx PHYs have the capability to provide 50MHz reference clock
in RMII mode and output on REF_CLK pin. Therefore, add the new property
"nxp,rmii-refclk-output" to support this feature. This property is only
available for PHYs which use nxp-c45-tja11xx driver, such as TJA1103,
TJA1104, TJA1120 and TJA1121.
====================
Wei Fang [Thu, 10 Oct 2024 06:19:44 +0000 (14:19 +0800)]
net: phy: c45-tja11xx: add support for outputting RMII reference clock
For TJA11xx PHYs, they have the capability to output 50MHz reference
clock on REF_CLK pin in RMII mode, which is called "revRMII" mode in
the PHY data sheet.
Per the RMII specification, the REF_CLK is sourced from MAC to PHY
or from an external source. But for TJA11xx PHYs, they support to
output a 50MHz RMII reference clock on REF_CLK pin. Previously the
"nxp,rmii-refclk-in" was added to indicate that in RMII mode, if
this property present, REF_CLK is input to the PHY, otherwise it
is output. This seems inappropriate now. Because according to the
RMII specification, the REF_CLK is originally input, so there is
no need to add an additional "nxp,rmii-refclk-in" property to
declare that REF_CLK is input.
Unfortunately, because the "nxp,rmii-refclk-in" property has been
added for a while, and we cannot confirm which DTS use the TJA1100
and TJA1101 PHYs, changing it to switch polarity will cause an ABI
break. But fortunately, this property is only valid for TJA1100 and
TJA1101. For TJA1103/TJA1104/TJA1120/TJA1121 PHYs, this property is
invalid because they use the nxp-c45-tja11xx driver, which is a
different driver from TJA1100/TJA1101. Therefore, for PHYs using
nxp-c45-tja11xx driver, add "nxp,rmii-refclk-out" property to
support outputting RMII reference clock on REF_CLK pin.
Jakub Kicinski [Fri, 11 Oct 2024 23:03:11 +0000 (16:03 -0700)]
selftests: net: move EXTRA_CLEAN of libynl.a into ynl.mk
Commit 1fd9e4f25782 ("selftests: make kselftest-clean remove libynl outputs")
added EXTRA_CLEAN of YNL generated files to ynl.mk. We already had
a EXTRA_CLEAN in the file including the snippet. Consolidate them.
Jakub Kicinski [Fri, 11 Oct 2024 23:03:10 +0000 (16:03 -0700)]
selftests: net: rebuild YNL if dependencies changed
Try to rebuild YNL if either user added a new family or the specs
of the families have changed. Stanislav's ncdevmem cause a false
positive build failure in NIPA because libynl.a isn't rebuilt
after ethtool is added to YNL_GENS.
Note that sha1sum is already used in other parts of the build system.
====================
Add support for per-NAPI config via netlink
Greetings:
Welcome to v6. Minor changes from v5 [1], please see changelog below.
There were no explicit comments from reviewers on the call outs in my
v5, so I'm retaining them from my previous cover letter just in case :)
A few important call outs for reviewers:
1. This revision seems to work (see below for a full walk through). I
think this is the behavior we talked about, but please let me know if
a use case is missing.
2. Re a previous point made by Stanislav regarding "taking over a NAPI
ID" when the channel count changes: mlx5 seems to call napi_disable
followed by netif_napi_del for the old queues and then calls
napi_enable for the new ones. In this RFC, the NAPI ID generation is
deferred to napi_enable. This means we won't end up with two of the
same NAPI IDs added to the hash at the same time.
Can we assume all drivers will napi_disable the old queues before
napi_enable the new ones?
- If yes: we might not need to worry about a NAPI ID takeover
function.
- If no: I'll need to make a change so that the NAPI ID generation
is deferred only for drivers which have opted into the config
space via calls to netif_napi_add_config
3. I made the decision to remove the WARN_ON_ONCE that (I think?)
Jakub previously suggested in alloc_netdev_mqs (WARN_ON_ONCE(txqs
!= rxqs);) because this was triggering on every kernel boot with my
mlx5 NIC.
4. I left the "maxqs = max(txqs, rxqs);" in alloc_netdev_mqs despite
thinking this is a bit strange. I think it's strange that we might
be short some number of NAPI configs, but it seems like most people
are in favor of this approach, so I've left it.
I'd appreciate thoughts from reviewers on the above items, if at all
possible.
Now, on to the implementation.
Firstly, this implementation moves certain settings to napi_struct so that
they are "per-NAPI", while taking care to respect existing sysfs
parameters which are interface wide and affect all NAPIs:
- NAPI ID
- gro_flush_timeout
- defer_hard_irqs
Furthermore:
- NAPI ID generation and addition to the hash is now deferred to
napi_enable, instead of during netif_napi_add
- NAPIs are removed from the hash during napi_disable, instead of
netif_napi_del.
- An array of "struct napi_config" is allocated in net_device.
IMPORTANT: The above changes affect all network drivers.
Optionally, drivers may opt-in to using their config space by calling
netif_napi_add_config instead of netif_napi_add.
If a driver does this, the NAPI being added is linked with an allocated
"struct napi_config" and the per-NAPI settings (including NAPI ID) are
persisted even as hardware queues are destroyed and recreated.
To help illustrate how this would end up working, I've added patches for
3 drivers, of which I have access to only 1:
- mlx5 which is the basis of the examples below
- mlx4 which has TX only NAPIs, just to highlight that case. I have
only compile tested this patch; I don't have this hardware.
- bnxt which I have only compiled tested. I don't have this
hardware.
NOTE: I only tested this on mlx5; I have no access to the other hardware
for which I provided patches. Hopefully other folks can help test :)
Here's how it works when I test it on my mlx5 system:
Joe Damato [Fri, 11 Oct 2024 18:45:00 +0000 (18:45 +0000)]
net: napi: Add napi_config
Add a persistent NAPI config area for NAPI configuration to the core.
Drivers opt-in to setting the persistent config for a NAPI by passing an
index when calling netif_napi_add_config.
napi_config is allocated in alloc_netdev_mqs, freed in free_netdev
(after the NAPIs are deleted).
Drivers which call netif_napi_add_config will have persistent per-NAPI
settings: NAPI IDs, gro_flush_timeout, and defer_hard_irq settings.
Per-NAPI settings are saved in napi_disable and restored in napi_enable.
Joe Damato [Fri, 11 Oct 2024 18:44:58 +0000 (18:44 +0000)]
net: napi: Make gro_flush_timeout per-NAPI
Allow per-NAPI gro_flush_timeout setting.
The existing sysfs parameter is respected; writes to sysfs will write to
all NAPI structs for the device and the net_device gro_flush_timeout
field. Reads from sysfs will read from the net_device field.
The ability to set gro_flush_timeout on specific NAPI instances will be
added in a later commit, via netdev-genl.
Joe Damato [Fri, 11 Oct 2024 18:44:56 +0000 (18:44 +0000)]
net: napi: Make napi_defer_hard_irqs per-NAPI
Add defer_hard_irqs to napi_struct in preparation for per-NAPI
settings.
The existing sysfs parameter is respected; writes to sysfs will write to
all NAPI structs for the device and the net_device defer_hard_irq field.
Reads from sysfs show the net_device field.
The ability to set defer_hard_irqs on specific NAPI instances will be
added in a later commit, via netdev-genl.
Daniel Golle [Fri, 11 Oct 2024 03:40:39 +0000 (04:40 +0100)]
net: phylink: allow half-duplex modes with RATE_MATCH_PAUSE
PHYs performing rate-matching using MAC-side flow-control always
perform duplex-matching as well in case they are supporting
half-duplex modes at all.
No longer remove half-duplex modes from their capabilities.
Simon Horman [Wed, 9 Oct 2024 09:40:10 +0000 (10:40 +0100)]
tg3: Address byte-order miss-matches
Address byte-order miss-matches flagged by Sparse.
In tg3_load_firmware_cpu() and tg3_get_device_address()
this is done using appropriate types to store big endian values.
In the cases of tg3_test_nvram(), where buf is an array which
contains values of several different types, cast to __le32
before converting values to host byte order.
Reported by Sparse as:
.../tg3.c:3745:34: warning: cast to restricted __be32
.../tg3.c:13096:21: warning: cast to restricted __le32
.../tg3.c:13096:21: warning: cast from restricted __be32
.../tg3.c:13101:21: warning: cast to restricted __le32
.../tg3.c:13101:21: warning: cast from restricted __be32
.../tg3.c:17070:63: warning: incorrect type in argument 3 (different base types)
.../tg3.c:17070:63: expected restricted __be32 [usertype] *val
.../tg3.c:17070:63: got unsigned int *
dr.../tg3.c:17071:63: warning: incorrect type in argument 3 (different base types)
.../tg3.c:17071:63: expected restricted __be32 [usertype] *val
.../tg3.c:17071:63: got unsigned int *
Also, address white-space issues on lines modified for the above.
And, for consistency, lines adjacent to them.
Compile tested only.
No functional change intended.
Currently xsk_cq_{reserve_addr,submit,cancel}_locked() take xdp_sock as
an input argument but it is only used for pulling out xsk_buff_pool
pointer from it.
Change mentioned functions to take pool pointer as an input argument to
avoid unnecessary dereferences.
Continue the process of dieting xdp_buff_xsk by removing orig_addr
member. It can be calculated from xdp->data_hard_start where it was
previously used, so it is not anything that has to be carried around in
struct used widely in hot path.
This has been used for initializing xdp_buff_xsk::frame_dma during pool
setup and as a shortcut in xp_get_handle() to retrieve address provided
to xsk Rx queue.
Let's bring xdp_buff_xsk back to occupying 2 cachelines by removing
xskb_list_node - for the purpose of gathering the xskb frags
free_list_node can be used, head of the list (xsk_buff_pool::xskb_list)
stays as-is, just reuse the node ptr.
It is safe to do as a single xdp_buff_xsk can never reside in two
pool's lists simultaneously.
David S. Miller [Mon, 14 Oct 2024 12:20:41 +0000 (13:20 +0100)]
Merge branch 'net-ti-ethernet-warnings'
Simon Horman says:
====================
net: ethernet: ti: Address some warnings
This patchset addresses some warnings flagged by Sparse, and clang-18 in
TI Ethernet drivers.
Although these changes do not alter the functionality of the code, by
addressing them real problems introduced in future which are flagged by
tooling will stand out more readily.
Compile tested only.
---
Changes in v2:
- Dropped patch to directly address __percpu Sparse warnings and, instead
- Add patch to use tstats
- Added tags
- Thanks to all for the review of v1
- Link to v1: https://lore.kernel.org/r/20240910-ti-warn-v1-0-afd1e404abbe@kernel.org
====================
Simon Horman [Thu, 10 Oct 2024 11:04:10 +0000 (12:04 +0100)]
net: ethernet: ti: am65-cpsw: Use __be64 type for id_temp
The id_temp local variable in am65_cpsw_nuss_probe() is
used to hold a 64-bit big-endian value as it is assigned using
cpu_to_be64().
It is read using memcpy(), where it is written as an identifier into a
byte-array. So this can also be treated as big endian.
As it's type is currently host byte order (u64), sparse flags
an endian mismatch when compiling for little-endian systems:
.../am65-cpsw-nuss.c:3454:17: warning: incorrect type in assignment (different base types)
.../am65-cpsw-nuss.c:3454:17: expected unsigned long long [usertype] id_temp
.../am65-cpsw-nuss.c:3454:17: got restricted __be64 [usertype]
Address this by using __be64 as the type of id_temp.
No functional change intended.
Compile tested only.
Heiner Kallweit [Thu, 10 Oct 2024 10:58:02 +0000 (12:58 +0200)]
r8169: enable SG/TSO on selected chip versions per default
Due to problem reports in the past SG and TSO/TSO6 are disabled per
default. It's not fully clear which chip versions are affected, so we
may impact also users of unaffected chip versions, unless they know
how to use ethtool for enabling SG/TSO/TSO6.
Vendor drivers r8168/r8125 enable SG/TSO/TSO6 for selected chip
versions per default, I'd interpret this as confirmation that these
chip versions are unaffected. So let's do the same here.
David S. Miller [Sun, 13 Oct 2024 17:02:50 +0000 (18:02 +0100)]
Merge branch 'ethtool-write-firmware'
Danielle Ratson says:
====================
ethtool: Add support for writing firmware
In the CMIS specification for pluggable modules, LPL (Local Payload) and
EPL (Extended Payload) are two types of data payloads used for managing
various functions and features of the module.
EPL payloads are used for more complex and extensive management functions
that require a larger amount of data, so writing firmware blocks using EPL
is much more efficient.
Currently, only LPL payload is supported for writing firmware blocks to
the module.
Add support for writing firmware block using EPL payload, both to support
modules that support only EPL write mechanism, and to optimize the flashing
process of modules that support LPL and EPL.
Running the flashing command on the same sample module using EPL vs. LPL
showed an improvement of 84%.
Patchset overview:
Patch #1: preparations
Patch #2: Add EPL support
v5: Resending- no changes.
v4: Resending the right version after wrong v3.
No changes from v2.
v2:
* Fix the commit meassges to align the cover letter about the
right meaning of LPL and EPL.
Patch #2:
* Initialize the variable 'bytes_written' before the first
iteration.
====================
Danielle Ratson [Wed, 9 Oct 2024 10:53:47 +0000 (13:53 +0300)]
net: ethtool: Add support for writing firmware blocks using EPL payload
In the CMIS specification for pluggable modules, LPL (Local Payload) and
EPL (Extended Payload) are two types of data payloads used for managing
various functions and features of the module.
EPL payloads are used for more complex and extensive management
functions that require a larger amount of data, so writing firmware
blocks using EPL is much more efficient.
Currently, only LPL payload is supported for writing firmware blocks to
the module.
Add support for writing firmware block using EPL payload, both to
support modules that supports only EPL write mechanism, and to optimize
the flashing process of modules that support LPL and EPL.
Danielle Ratson [Wed, 9 Oct 2024 10:53:46 +0000 (13:53 +0300)]
net: ethtool: Add new parameters and a function to support EPL
In the CMIS specification for pluggable modules, LPL (Local Payload) and
EPL (Extended Payload) are two types of data payloads used for managing
various functions and features of the module.
EPL payloads are used for more complex and extensive management
functions that require a larger amount of data, so writing firmware
blocks using EPL is much more efficient.
Currently, only LPL payload is supported for writing firmware blocks to
the module.
Add EPL related parameters to the function ethtool_cmis_cdb_compose_args()
and add a specific function for calculating the maximum allowable length
extension for EPL. Both will be used in the next patch to add support for
writing firmware blocks using EPL.
Changes since v6:
- fix some typos in the document for SKB_DROP_REASON_TUNNEL_TXINFO
Changes since v5:
- fix some typos in the document for SKB_DROP_REASON_TUNNEL_TXINFO
Changes since v4:
- make skb_vlan_inet_prepare() return drop reasons, instead of introduce
a wrapper for it in the 3rd patch.
- modify the document for SKB_DROP_REASON_LOCAL_MAC and
SKB_DROP_REASON_TUNNEL_TXINFO.
Changes since v3:
- rename SKB_DROP_REASON_VXLAN_INVALID_SMAC to
SKB_DROP_REASON_MAC_INVALID_SOURCE in the 6th patch
Changes since v2:
- move all the drop reasons of VXLAN to the "core", instead of introducing
the VXLAN drop reason subsystem
- add the 6th patch, which capture the drop reasons from vxlan_snoop()
- move the commits for vxlan_remcsum() and vxlan_set_mac() after
vxlan_rcv() to update the call of them accordingly
- fix some format problems
Changes since v1:
- document all the drop reasons that we introduce
- rename the drop reasons to make them more descriptive, as Ido advised
- remove the 2nd patch, which introduce the SKB_DR_RESET
- add the 4th patch, which adds skb_vlan_inet_prepare_reason() helper
- introduce the 6th patch, which make vxlan_set_mac return drop reasons
- introduce the 10th patch, which uses VXLAN_DROP_NO_REMOTE as the drop
reasons, as Ido advised
====================
Menglong Dong [Wed, 9 Oct 2024 02:28:27 +0000 (10:28 +0800)]
net: vxlan: add drop reasons support to vxlan_xmit_one()
Replace kfree_skb/dev_kfree_skb with kfree_skb_reason in vxlan_xmit_one.
No drop reasons are introduced in this commit.
The only concern of mine is replacing dev_kfree_skb with
kfree_skb_reason. The dev_kfree_skb is equal to consume_skb, and I'm not
sure if we can change it to kfree_skb here. In my option, the skb is
"dropped" here, isn't it?
Menglong Dong [Wed, 9 Oct 2024 02:28:26 +0000 (10:28 +0800)]
net: vxlan: use kfree_skb_reason() in vxlan_xmit()
Replace kfree_skb() with kfree_skb_reason() in vxlan_xmit(). Following
new skb drop reasons are introduced for vxlan:
/* no remote found for xmit */
SKB_DROP_REASON_VXLAN_NO_REMOTE
/* packet without necessary metadata reached a device which is
* in "external" mode
*/
SKB_DROP_REASON_TUNNEL_TXINFO
Menglong Dong [Wed, 9 Oct 2024 02:28:25 +0000 (10:28 +0800)]
net: vxlan: make vxlan_set_mac() return drop reasons
Change the return type of vxlan_set_mac() from bool to enum
skb_drop_reason. In this commit, the drop reason
"SKB_DROP_REASON_LOCAL_MAC" is introduced for the case that the source
mac of the packet is a local mac.
Introduce the function pskb_inet_may_pull_reason() and make
pskb_inet_may_pull a simple inline call to it. The drop reasons of it just
come from pskb_may_pull_reason().
Introduce the function pskb_network_may_pull_reason() and make
pskb_network_may_pull() a simple inline call to it. The drop reasons of
it just come from pskb_may_pull_reason.
====================
microchip_t1s: Update on Microchip 10BASE-T1S PHY driver
This patch series contain the below updates:
- Restructured lan865x_write_cfg_params() and lan865x_read_cfg_params()
functions arguments to more generic.
- Updated new/improved initial settings of LAN865X Rev.B0 from latest
AN1760.
- Added support for LAN865X Rev.B1 from latest AN1760.
- Moved LAN867X reset handling to a new function for flexibility.
- Added support for LAN867X Rev.C1/C2 from latest AN1699.
- Disabled/enabled collision detection based on PLCA setting.
====================
net: phy: microchip_t1s: configure collision detection based on PLCA mode
As per LAN8650/1 Rev.B0/B1 AN1760 (Revision F (DS60001760G - June 2024))
and LAN8670/1/2 Rev.C1/C2 AN1699 (Revision E (DS60001699F - June 2024)),
under normal operation, the device should be operated in PLCA mode.
Disabling collision detection is recommended to allow the device to
operate in noisy environments or when reflections and other inherent
transmission line distortion cause poor signal quality. Collision
detection must be re-enabled if the device is configured to operate in
CSMA/CD mode.
net: phy: microchip_t1s: add support for Microchip's LAN867X Rev.C2
Add support for LAN8670/1/2 Rev.C2 as per the latest configuration note
AN1699 released (Revision E (DS60001699F - June 2024)) for Rev.C1 is also
applicable for Rev.C2. Refer hardware revisions list in the latest AN1699
Revision E (DS60001699F - June 2024).
https://www.microchip.com/en-us/application-notes/an1699
net: phy: microchip_t1s: add support for Microchip's LAN867X Rev.C1
Add support for LAN8670/1/2 Rev.C1 as per the latest configuration note
AN1699 released (Revision E (DS60001699F - June 2024)).
https://www.microchip.com/en-us/application-notes/an1699
net: phy: microchip_t1s: move LAN867X reset handling to a new function
Move LAN867X reset handling code to a new function called
lan867x_check_reset_complete() which will be useful for the next patch
which also uses the same code to handle the reset functionality.
net: phy: microchip_t1s: add support for Microchip's LAN865X Rev.B1
Add support for LAN8650/1 Rev.B1. As per the latest configuration note
AN1760 released (Revision F (DS60001760G - June 2024)) for Rev.B0 is also
applicable for Rev.B1. Refer hardware revisions list in the latest AN1760
Revision F (DS60001760G - June 2024).
https://www.microchip.com/en-us/application-notes/an1760
net: phy: microchip_t1s: update new initial settings for LAN865X Rev.B0
Update the new/improved initial settings from the latest configuration
application note AN1760 released for LAN8650/1 Rev.B0 Revision F
(DS60001760G - June 2024).
https://www.microchip.com/en-us/application-notes/an1760
Restructure lan865x_write_cfg_params() and lan865x_read_cfg_params()
functions arguments to more generic which will be useful for the next
patch which updates the improved initial configuration for LAN8650/1
Rev.B0 published in the Configuration Note.
Commit b3ea416419c8 ("testing: net-drv: add basic shaper test")
removed the trailing backslash from the last entry. We have
a terminating comment here to avoid having to modify the last
line when adding at the end.
Hangbin Liu [Thu, 10 Oct 2024 04:00:27 +0000 (04:00 +0000)]
selftests: rtnetlink: update netdevsim ipsec output format
After the netdevsim update to use human-readable IP address formats for
IPsec, we can now use the source and destination IPs directly in testing.
Here is the result:
# ./rtnetlink.sh -t kci_test_ipsec_offload
PASS: ipsec_offload
Hangbin Liu [Thu, 10 Oct 2024 04:00:26 +0000 (04:00 +0000)]
netdevsim: copy addresses for both in and out paths
The current code only copies the address for the in path, leaving the out
path address set to 0. This patch corrects the issue by copying the addresses
for both the in and out paths. Before this patch:
====================
net: xilinx: emaclite: Adopt clock support
This patchset adds emaclite clock support. AXI Ethernet Lite IP can also
be used on SoC platforms like Zynq UltraScale+ MPSoC which combines
powerful processing system (PS) and user-programmable logic (PL) into
the same device. On these platforms it is mandatory to explicitly enable
IP clocks for proper functionality.
====================
Abin Joseph [Wed, 9 Oct 2024 16:28:21 +0000 (21:58 +0530)]
dt-bindings: net: emaclite: Add clock support
Add s_axi_aclk AXI4 clock support. Traditionally this IP was used on
microblaze platforms which had fixed clocks enabled all the time. But
since its a PL IP, it can also be used on SoC platforms like Zynq
UltraScale+ MPSoC which combines processing system (PS) and user
programmable logic (PL) into the same device. On these platforms instead
of fixed enabled clocks it is mandatory to explicitly enable IP clocks
for proper functionality.
So make clock a required property and also define max supported clock
constraints.
Aryan Srivastava [Thu, 10 Oct 2024 00:49:34 +0000 (13:49 +1300)]
net: phy: aquantia: poll status register
The system interface connection status register is not immediately
correct upon line side link up. This results in the status being read as
OFF and then transitioning to the correct host side link mode with a
short delay. This causes the phylink framework passing the OFF status
down to all MAC config drivers, resulting in the host side link being
misconfigured, which in turn can lead to link flapping or complete
packet loss in some cases.
Mitigate this by periodically polling the register until it not showing
the OFF state. This will be done every 1ms for 10ms, using the same
poll/timeout as the processor intensive operation reads.
If the phy is still expressing the OFF state after the timeout, then set
the link to false and pass the NA interface mode onto the phylink
framework.
Jakub Kicinski [Tue, 8 Oct 2024 15:48:24 +0000 (08:48 -0700)]
eth: remove the DLink/Sundance (ST201) driver
Konstantin reports the maintainer's address bounces.
There is no other maintainer and the driver is quite old.
There is a good chance nobody is using this driver any more.
Let's try to remove it completely, we can revert it back in
if someone complains.
Jakub Kicinski [Fri, 11 Oct 2024 01:40:34 +0000 (18:40 -0700)]
Merge branch 'tg3-link-irqs-napis-and-queues'
Joe Damato says:
====================
tg3: Link IRQs, NAPIs, and queues
This follows from a previous RFC (wherein I botched the subject lines of
all the messages) [1].
I've taken Michael Chan's suggestion on modifying patch 2 and I've
updated the commit messages of both patches to test and show the output
for the default 1 TX 4 RX queues and the 4 TX and 4 RX queues cases.
Reviewers: please check the commit messages carefully to ensure the
output is correct (or on your own systems to verify, if you like). I am
not a tg3 expert and it's possible that I got something wrong.
Examine /proc/interrupts once again, noting that tg3 will now rename the
IRQs to suggest that they are combined tx and rx without allocating
additional IRQs, so the total IRQ count in /proc/interrupts is
unchanged:
Heiner Kallweit [Wed, 9 Oct 2024 05:48:05 +0000 (07:48 +0200)]
r8169: remove original workaround for RTL8125 broken rx issue
Now that we have b9c7ac4fe22c ("r8169: disable ALDPS per default for
RTL8125"), the first attempt to fix the issue shouldn't be needed
any longer. So let's effectively revert 621735f59064 ("r8169: fix
rare issue with broken rx after link-down on RTL8125") and see
whether anybody complains.
Martin KaFai Lau [Fri, 11 Oct 2024 00:53:55 +0000 (17:53 -0700)]
Merge branch 'selftests/bpf: add coverage for xdp_features in test_progs'
Alexis Lothoré says:
====================
this small series aims to increase coverage of xdp features in
test_progs. The initial versions proposed to rework test_xdp_features.sh
to make it fit in test_progs, but some discussions in v1 and v2 showed
that the script is still needed as a standalone tool. So this new
revision lets test_xdp_features.sh as-is, and rather adds missing
coverage in existing test (cpu map). The new revision is now also a
follow-up to the update performed by Florian Kauer in [1] for devmap
programs testing.
[1] https://lore.kernel.org/bpf/20240911-devel-koalo-fix-ingress-ifindex-v4-2-5c643ae10258@linutronix.de/
---
Changes in v3:
- Drop xdp_features rework commit
- update xdp_cpumap_attach to extend its coverage
- Link to v2: https://lore.kernel.org/r/20240910-convert_xdp_tests-v2-1-a46367c9d038@bootlin.com
Changes in v2:
- fix endianness management in userspace packet parsing (call htonl on
constant rather than packet part)
The new test has been run in a local x86 environment and in CI:
#560/1 xdp_cpumap_attach/CPUMAP with programs in entries:OK
#560/2 xdp_cpumap_attach/CPUMAP with frags programs in entries:OK
#560 xdp_cpumap_attach:OK
Summary: 1/2 PASSED, 0 SKIPPED, 0 FAILED
====================
selftests/bpf: check program redirect in xdp_cpumap_attach
xdp_cpumap_attach, in its current form, only checks that an xdp cpumap
program can be executed, but not that it performs correctly the cpu
redirect as configured by userspace (bpf_prog_test_run_opts will return
success even if the redirect program returns an error)
Add a check to ensure that the program performs the configured redirect
as well. The check is based on a global variable incremented by a
chained program executed only if the redirect program properly executes.
selftests/bpf: make xdp_cpumap_attach keep redirect prog attached
Current test only checks attach/detach on cpu map type program, and so
does not check that it can be properly executed, neither that it
redirects correctly.
Update the existing test to extend its coverage:
- keep the redirected program loaded
- try to execute it through bpf_prog_test_run_opts with some dummy
context
While at it, bring the following minor improvements:
- isolate test interface in its own namespace
selftests/bpf: fix bpf_map_redirect call for cpu map test
xdp_redir_prog currently redirects packets based on the entry at index 1
in cpu_map, but the corresponding test only manipulates the entry at
index 0. This does not really affect the test in its current form since
the program is detached before having the opportunity to execute, but it
needs to be fixed before being able improve the corresponding test (ie,
not only test attach/detach but also the redirect feature)
Fix this XDP program by making it redirect packets based on entry 0 in
cpu_map instead of entry 1.
projects / linux.git / log